JP7422198B1 - Leading indicator calculation device, control method and program for leading indicator calculation device - Google Patents

Abstract

[Problem] To calculate a more accurate and more advanced indicator of office rental market conditions. [Solution] A leading indicator calculation device that calculates leading indicators of office rental market conditions, the market data acquisition unit acquiring aggregated data for each predetermined period of multiple types of market data that change according to trends in the office rental market. and a preprocessing execution unit that performs predetermined preprocessing on at least some of the market condition data so that at least some of the market condition data changes in advance of changes in office rental market conditions; For each of the market data, aggregate data for each period is converted into a score according to the degree of rarity within the market data, and a representative value for each period is determined from the scores for each period of the plurality of types of market data. A leading indicator calculation unit that calculates the leading indicator. [Selection diagram] Figure 21

Description

The present invention relates to a leading indicator calculation device, a control method for the leading indicator calculation device, and a program.

Office rental market conditions, which indicate business confidence in the rental office building market, are important as one of the factors for determining the economic situation. For this reason, various reports and data regarding office rental market conditions are published every day from various sources.

These reports and data often analyze the office rental market by focusing on the vacancy rate of rental office buildings, but they also analyze the office rental market by focusing on asking rents and the number of new office building construction starts. In some cases, Furthermore, when predicting the future of the office rental market, decisions are often made by looking at trends in data from the past to the present (upward trends, downward trends, etc.).

Under these circumstances, a technology has been developed for predicting shipping market conditions, which fluctuate under the influence of the economy in the same way as office rental market conditions (see, for example, Patent Document 1).

JP 2019-179354 Publication

However, when forecasting the office rental market by looking at trends in data such as vacancy rates, it is difficult to foresee a change in trends, and it is also difficult to forecast over the long term.

On the other hand, in some cases, such as a new office building development project, the project may take a long time from start to finish, and there is a need to predict future trends in the office building rental market at an earlier stage. The same is true for investors investing in office buildings and tenants in office buildings, and it is important to forecast the future of the office rental market from a long-term perspective, including changes in trends.

The present invention has been made in view of these problems, and provides a leading indicator calculation device, a method for controlling the leading indicator calculation device, and a program that can calculate a leading indicator of office rental market conditions that is more accurate and more proactive. The purpose is to

A leading indicator calculation device according to an embodiment of the present invention is a leading indicator calculation device that calculates a leading indicator of office rental market conditions, and is configured to collect multiple types of market data that change according to trends in the office rental market every predetermined period. a market condition data acquisition unit that acquires aggregated data of the office rental market ; a preprocessing execution unit that performs predetermined preprocessing on the data; and a leading indicator calculation unit that calculates, as the leading indicator, a representative value for each period found from the scores for each period of the market data.

Other problems disclosed by the present application and methods for solving the problems will be made clear by the description in the Detailed Description section and the drawings.

It becomes possible to calculate more accurate and more advanced indicators of office rental market conditions.

FIG. 2 is a diagram showing the hardware configuration of a leading indicator calculation device. FIG. 3 is a diagram showing a storage device of the leading indicator calculation device. FIG. 3 is a diagram showing a first office data management table. It is a figure which shows the 2nd office data management table. It is a figure which shows the 3rd office data management table. It is a figure which shows the 4th office data management table. It is a figure which shows the 5th office data management table. It is a figure which shows the 6th office data management table. FIG. 3 is a diagram showing a leading indicator management table. FIG. 3 is a diagram illustrating an example of a leading indicator calculated by a leading indicator calculation device. FIG. 3 is a diagram showing a selection process of office data. FIG. 7 is a diagram showing verification results of preprocessing performed on each office data. FIG. 7 is a diagram showing the influence of performing preprocessing on each office data. FIG. 6 is a diagram illustrating an example of verifying preprocessing for first office data. FIG. 7 is a diagram illustrating an example of verifying preprocessing for second office data. FIG. 7 is a diagram illustrating an example of verifying preprocessing for third office data. It is a figure which shows the example of verification of the preprocessing with respect to 4th office data. It is a figure which shows the verification example of the preprocessing with respect to 5th office data. It is a figure which shows the example of verification of the preprocessing with respect to 6th office data. FIG. 2 is a diagram showing a functional configuration of a leading indicator calculation device. FIG. 3 is a diagram illustrating an example of market condition analysis using leading indicators calculated by a leading indicator calculation device. FIG. 2 is a diagram illustrating how market conditions are judged using leading indicators calculated by a leading indicator calculation device. FIG. 3 is a diagram illustrating an example of market condition analysis using leading indicators calculated by a leading indicator calculation device. FIG. 3 is a diagram illustrating an example of verifying a leading indicator calculated by a leading indicator calculation device. It is a figure which shows the flowchart which shows the flow of processing by a leading indicator calculation device.

From the description of this specification and the attached drawings, at least the following matters will become clear. DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below based on one embodiment thereof with reference to the accompanying drawings.

== Leading indicator calculation device ==
FIG. 1 shows the hardware configuration of a leading indicator calculation device 200 according to an embodiment of the present invention. The leading indicator calculation device 200 is an information processing device such as a personal computer, server, smartphone, tablet, etc. that calculates a leading indicator of office rental market conditions (hereinafter also referred to as Office-RISM).

The leading indicator calculation device 200 uses a plurality of types (six types in this embodiment) of market condition data regarding office rental market conditions that are aggregated for each predetermined period (every three months in this embodiment). Calculate leading indicators of market conditions (Office-RISM).

Details of the creation procedure, accuracy, and leading nature of this leading indicator will be described later, but for example, as shown in Figure 21, the city center 5 Compared to trends in average asking rents for the wards (Chiyoda, Chuo, Minato, Shinjuku, and Shibuya wards, hereinafter also referred to as designated areas), it can be confirmed that there is almost exactly a two to five year lead.

In this manner, the leading indicator calculation device 200 according to the present embodiment makes it possible to calculate a more accurate and more advanced leading indicator of office rental market conditions (Office-RISM).

<Hardware configuration>
Returning to FIG. 1, the leading indicator calculation device 200 includes a CPU (Central Processing Unit) 210, a memory 220, a communication device 230, a storage device 240, an input device 250, an output device 260, and a recording medium reading device 270. be done.

The storage device 240 stores various programs and data executed by the CPU 210. These programs and data are read into the memory 220 and executed or processed by the CPU 210, thereby realizing various functions of the leading indicator calculation device 200.

Here, the storage device 240 is a nonvolatile storage device such as a hard disk, SSD (Solid State Drive), or flash memory.

As shown in FIG. 2, the storage device 240 includes a leading indicator calculation device control program 700, a first office data management table 610, a second office data management table 620, a third office data management table 630, and a fourth office data management table. 640, a fifth office data management table 650, a sixth office data management table 660, and a leading indicator management table 670 are stored.

The recording medium reading device 270 reads the leading indicator calculation device control program 700 and data recorded on a recording medium 800 such as a CD, DVD, or SD card, and stores them in the storage device 240.

The communication device 230 is communicably connected to other information processing devices (not shown) via the network 500, and exchanges the leading indicator calculation device control program 700 and various data. For example, the above-mentioned leading indicator calculation device control program 700 and various data may be stored in another computer, and the leading indicator calculation device 200 may download the leading indicator calculation device control program 700 and data from this computer. .

The input device 250 is a device such as various buttons, switches, a mouse, a keyboard, etc. that accepts input of commands and data by the operator of the leading indicator calculation device 200.

The output device 260 is, for example, a display device such as a display, or a device such as a speaker.

Further, the network 500 is various information communication networks such as the Internet, LAN (Local Area Network), and telephone network.

Note that the leading indicator calculation device 200 may be configured with one information processing device or may be configured with a plurality of information processing devices. Further, the leading indicator calculation device 200 may be a virtual machine or a cloud computer.

<First office data management table>
The first office data management table 610 stored in the storage device 240 is shown in FIG. The first office data management table 610 is a table that stores first office data (also referred to as first market data), which is one of the six market data used to calculate Office-RISM.

The first office data consists of, for example, aggregated data of the percentage of rental office buildings in the five central wards (prescribed area) that are fully occupied (occupancy rate) and are compiled every three months. The full occupancy rate of rental office buildings is calculated by subtracting the ratio (%) of rental office buildings with vacant rooms published by Miki Shoji Co., Ltd. from 100 (%).

The full occupancy rate of a rental office building is market data regarding the operating status of rental offices, and the value tends to increase as the office rental market becomes more favorable. For this reason, the full occupancy rate of rental office buildings tends to increase or decrease in line with trends in the office rental market.

Column (A) lists the last day of each three-month period. In the example shown in Figure 3, the earliest period is the first quarter of 2000 (the last day is March 31, 2000), and the latest period is the first quarter of 2022 (the last day is March 31, 2022). It is.

Column (B) contains aggregated data on the ratio (%) of rental office buildings with vacant rooms in the five central wards of Tokyo (designated areas) published by Miki Shoji Co., Ltd. every three months. Are listed.

Column (C) contains aggregated data on the occupancy rate of rental office buildings within the five central wards of Tokyo (designated areas) every three months by subtracting the value in column (B) from 100. There is.

Column (D) lists values obtained by replacing each aggregated data listed in column (C) with the amount of change from the aggregated data a certain period of time ago (four quarters ago in this embodiment).

Column (E) contains the value (m1 - σ1) obtained by subtracting 1.0 x standard deviation (σ1) from the average value (m1) of the aggregated data listed in column (D) (z value = -1). In the example shown in FIG. 3, it is "-7.9". Therefore, if the aggregated data listed in column (D) is smaller than "-7.9", the aggregated data statistically exceeds one standard deviation (σ1) from the average value (m1). You can see that it is small. In addition, as described in the (L) column and (M) column of FIG. 3, the average value (m1) of the aggregated data is 0.09, and the standard deviation (σ1) is 7.96.

Column (F) lists the value (m1 - 0.5 x σ1) obtained by subtracting 0.5 x standard deviation from the average value of the aggregated data listed in column (D) (z value = - (equivalent to 0.5).

In column (G), the average value (m1) of the aggregated data written in column (D) is written (corresponding to z value = 0).

In column (H), the value (m1 + 0.5 x σ1), which is the average value of the aggregated data written in column (D) plus 0.5 x standard deviation, is written (z value = 0.5 ).

In column (I), a value (m1+σ1) obtained by adding 1.0×standard deviation to the average value of the aggregated data written in column (D) is written (corresponding to z value=1).

Column (J) contains the score obtained by normalizing the aggregated data listed in column (D) according to the degree of rarity within the market data (first office data). ing.

In this embodiment, if the value of the aggregated data is less than "m1 - σ1" (z value is less than -1.0), the score is set to 0, and if the value of the aggregated data is less than "m1 - σ1" or more than "m1 - 0.5 x σ1", the score is set to 0. (If the z value is -1.0 or more and less than -0.5), the score = 20, and if the z value is more than "m1 - 0.5 x σ1" and less than "m1" (the z value is -0.5 or more and less than 0). If the score is "m1" or more and less than "m1 + 0.5 x σ1" (z value is 0 or more and less than + 0.5), the score = 60, and if the value is "m1 + 0.5 x σ1" or more and less than "m1 + σ1" When the z value is +0.5 or more and less than +1, the score is set as 80, and when the z value is equal to or larger than "m1+σ1" (the z value is +1.0 or more), the score is set as 100.

With this aspect, each aggregated data in column (D) can be normalized to a range of 0 to 100 points depending on the degree of rarity.

These scores indicate the strength of the office rental market, either upward or downward. For example, a period in which the first office data score is 100 points means that the aggregated data (occupancy rate) for that period is statistically a large value with a rarity of one standard deviation or more. This suggests that the strength of the rise in market conditions is that strong. On the other hand, a period in which the first office data score is 0 points means that the aggregated data for that period is statistically rare and small by one standard deviation or more, so the office rental market declines. It suggests that the strength to move towards is that much stronger.

Column (K) is a compilation of column (J) in one column.
<Second office data management table>

Next, FIG. 4 shows the second office data management table 620 stored in the storage device 240. The second office data management table 620 is a table that stores second office data (also referred to as second market data), which is one of the six market data used to calculate Office-RISM.

The second office data consists of, for example, aggregated data published by Sanko Estate Co., Ltd., which is compiled every three months of the net absorption of rental office buildings in the five central wards (prescribed areas) of Tokyo. Net absorption is obtained by calculating for each period the ``Available area at the beginning of the period for completed buildings + New supply area (total leased area of new buildings completed during the period) - Adopted area at the end of the period for completed buildings.'' It is a value that represents the increase or decrease in office space demand within a period.

Net absorption is market data regarding the operating status of rental offices, and its value tends to increase as the office rental market becomes more favorable. For this reason, net absorption tends to increase or decrease in line with trends in the office rental market.

Column (A) lists the last day of each three-month period. In the example shown in Figure 4, the earliest period is the first quarter of 2000 (the last day is March 31, 2000), and the latest period is the first quarter of 2022 (the last day is March 31, 2022). It is.

Column (B) contains aggregated data published by Sanko Estate Co., Ltd. on net absorption in the five central wards of Tokyo (designated areas), compiled every three months.

In column (C), values obtained by replacing each aggregated data listed in column (B) with a predetermined number (two in this embodiment) of backward moving average values are listed. In other words, each aggregated data in column (C) is a 6-month backward moving average of the aggregated data in column (B).

Column (D) lists the value (m2 - σ2) obtained by subtracting 1.0 x standard deviation (σ2) from the average value (m2) of the aggregated data listed in column (C) (z value = -1). In the example shown in FIG. 4, it is "-13604". Therefore, if the aggregated data listed in column (C) is smaller than "-13604", the aggregated data is statistically smaller than one standard deviation (σ2) from the average value (m2). I understand that. Note that, as described in columns (K) and (M) of FIG. 4, the average value (m2) of the aggregated data is 42,499, and the standard deviation (σ2) is 56,103.

Column (E) lists the value (m2 - 0.5 x σ2) obtained by subtracting 0.5 x standard deviation from the average value of the aggregated data listed in column (C) (z value = - (equivalent to 0.5).

In column (F), the average value (m2) of the aggregated data written in column (C) is listed (corresponding to z value = 0).

In column (G), the value (m2 + 0.5 x σ2), which is the average value of the aggregated data listed in column (C) plus 0.5 x standard deviation, is listed (z value = 0.5 ).

In column (H), a value (m2+σ2) obtained by adding 1.0 x standard deviation to the average value of the aggregated data written in column (C) is written (corresponding to z value = 1).

Column (I) contains the score obtained by normalizing the aggregated data listed in column (C) according to the degree of rarity within the relevant market data (within the second office data). ing.

In this embodiment, if the value of the aggregated data is less than "m2 - σ2" (z value is less than -1.0), the score is set to 0, and if the value of the aggregated data is less than "m2 - σ2" or more than "m2 - 0.5 x σ2", the score is set to 0. (z value is -1.0 or more and less than -0.5), score = 20, and score is "m2 - 0.5 x σ2" or more and less than "m2" (z value is -0.5 or more and less than 0). If the score is "m2" or more and less than "m2 + 0.5 x σ2" (z value is 0 or more and less than + 0.5), the score = 60, and if the value is "m2 + 0.5 x σ2" or more and less than "m2 + σ2" When the z value is +0.5 or more and less than +1, the score is set as 80, and when the z value is ``m2+σ2'' or more (the z value is +1.0 or more), the score is set as 100.

With this aspect, each aggregated data in column (C) can be normalized to a range of 0 to 100 points depending on the degree of rarity.

As with the primary office data, these scores indicate the strength of the upward or downward trend in office rental market conditions. For example, a period in which the second office data score is 100 points means that the aggregate data (net absorption) for that period is statistically rare and large by one standard deviation or more. This suggests that the strength of the rise in office rental market conditions is that strong. On the other hand, a period in which the second office data score is 0 points means that the aggregated data for that period is statistically rare and small by one standard deviation or more, so the office rental market declines. It suggests that the strength to move towards is that much stronger.

Column (J) is a compilation of column (I) in one column.
<Third office data management table>

Next, FIG. 5 shows the third office data management table 630 stored in the storage device 240. The third office data management table 630 is a table that stores third office data (also referred to as third market data), which is one of the six market data used to calculate Office-RISM.

The third office data consists of, for example, aggregated data published by Miki Shoji Co., Ltd. of the vacancy rates of rental office buildings in the five central wards (prescribed areas) of Tokyo, compiled every three months.

The vacancy rate is market data regarding the occupancy status of rental offices, and the value tends to decrease as the office rental market becomes more favorable. For this reason, vacancy rates tend to increase or decrease in a manner that goes against trends in the office rental market.

Column (A) lists the last day of each three-month period. In the example shown in Figure 5, the earliest period is the first quarter of 2000 (the last day is March 31, 2000), and the latest period is the first quarter of 2022 (the last day is March 31, 2022). It is.

Column (B) contains aggregated data published by Miki Shoji Co., Ltd. on the vacancy rate of rental office buildings in the five central wards of Tokyo (designated areas), compiled every three months.

Column (C) lists values obtained by replacing each aggregated data listed in column (B) with the amount of change from the aggregated data a certain period of time ago (four quarters ago in this embodiment).

Column (D) contains the value (m3 - σ3) obtained by subtracting 1.0 x standard deviation (σ3) from the average value (m3) of the aggregated data listed in column (C) (z value = -1). In the example shown in FIG. 5, it is "-1.7". Therefore, if the aggregated data listed in column (C) is smaller than "-1.7", the aggregated data statistically exceeds one standard deviation (σ3) from the average value (m3). You can see that it is small. Note that, as described in columns (L) and (M) of FIG. 5, the average value (m3) of the aggregated data is 0.047, and the standard deviation (σ3) is 1.76.

Column (E) lists the value (m3 - 0.5 x σ3) obtained by subtracting 0.5 x standard deviation from the average value of the aggregated data listed in column (C) (z value = - (equivalent to 0.5).

Column (F) lists the average value (m3) of the aggregated data listed in column (C) (corresponds to z value = 0).

In column (G), the value (m3 + 0.5 x σ3), which is the average value of the aggregated data listed in column (C) plus 0.5 x standard deviation, is listed (z value = 0.5 ).

Column (H) lists the value (m3+σ3) obtained by adding 1.0× standard deviation to the average value of the aggregated data listed in column (C) (equivalent to z value=1).

Column (I) contains the score obtained by normalizing the aggregated data listed in column (C) according to the degree of rarity within the relevant market data (within the third office data). ing.

In this embodiment, if the value of the aggregated data is less than "m3 - σ3" (z value is less than -1.0), the score = 100, and if the value is "m3 - σ3" or more and less than "m3 - 0.5 x σ3", the score is set as 100. (z value is -1.0 or more and less than -0.5), score = 80, and score is "m3 - 0.5 x σ3" or more and less than "m3" (z value is -0.5 or more and less than 0). If the score is "m3" or more and less than "m3 + 0.5 x σ3" (z value is 0 or more and less than + 0.5) then the score = 40, and if the value is "m3 + 0.5 x σ3" or more and less than "m3 + σ3" (The z value is +0.5 or more and less than +1), the score is set as 20, and the case that the z value is "m3+σ3" or more (the z value is +1.0 or more), the score is set as 0.

With this aspect, each aggregated data in column (C) can be normalized to a range of 0 to 100 points depending on the degree of rarity.

As with the primary office data, these scores indicate the strength of the upward or downward trend in office rental market conditions. For example, a period in which the score of the third office data is 100 points means that the aggregated data (vacancy rate) for that period is statistically rare and small by more than one standard deviation. This suggests that the strength of the rise in rental market conditions is that strong. On the other hand, a period in which the third office data score is 0 points means that the aggregated data for that period is statistically unusual and large by one standard deviation or more, so the office rental market declines. It suggests that the strength to move towards is that much stronger.

Column (J) is a compilation of column (I) in one column.
<4th office data management table>

Next, FIG. 6 shows a fourth office data management table 640 stored in the storage device 240. The fourth office data management table 640 is a table that stores fourth office data (also referred to as fourth market data), which is one of the six market data used to calculate Office-RISM.

The 4th office data is, for example, a building rental index that is part of the real estate industry business index published every three months, created based on a questionnaire survey of real estate industry managers conducted by the Land Research Institute, a general incorporated foundation. It consists of aggregated data compiled every three months from an index that shows the business outlook for the next three months (hereinafter also referred to as the Building Rental Business Conditions Index).

The building rental industry index is market data regarding business confidence held by rental office managers, and its value tends to increase as the office rental market becomes more favorable. Therefore, the building rental business index tends to increase or decrease in line with trends in the office rental market.

Column (A) lists the last day of each three-month period. In the example shown in Figure 6, the earliest period is the first quarter of 2000 (the last day is March 31, 2000), and the latest period is the first quarter of 2022 (the last day is March 31, 2022). It is.

Column (B) contains aggregated data on the building rental industry index compiled every three months.

Note that the real estate industry business index of the Land Research Institute, General Incorporated Foundation, is published on the first day of the quarter following the survey, but in this embodiment, the index is closer to the time when managers judge business confidence. In order to do so, the data for the quarter in which the questionnaire was conducted is listed in column (B).

Column (C) lists the value (m4 - σ4) obtained by subtracting 1.0 x standard deviation (σ4) from the average value (m4) of the aggregated data listed in column (B) (z value = -1). In the example shown in FIG. 6, it is "-26.2". Therefore, if the aggregated data listed in column (B) is smaller than "-26.2", the aggregated data statistically exceeds one standard deviation (σ4) from the average value (m4). You can see that it is small. Note that, as shown in columns (J) and (K) of FIG. 6, the average value (m4) of the aggregated data is -8.8, and the standard deviation (σ4) is 17.34.

Column (D) lists the value (m4 - 0.5 x σ4) obtained by subtracting 0.5 x standard deviation from the average value of the aggregated data listed in column (B) (z value = - (equivalent to 0.5).

In column (E), the average value (m4) of the aggregated data written in column (B) is listed (corresponding to z value = 0).

In column (F), the value (m4 + 0.5 x σ4), which is the average value of the aggregated data listed in column (B) plus 0.5 x standard deviation, is listed (z value = 0.5 ).

In column (G), a value (m4+σ4) obtained by adding 1.0× standard deviation to the average value of the aggregated data described in column (B) is written (corresponding to z value = 1).

In Column (H), the score obtained by normalizing the aggregated data listed in Column (B) according to the degree of rarity within the relevant market data (within the 4th office data) is entered. ing.

In this embodiment, if the value of the aggregated data is less than "m4 - σ4" (z value is less than -1.0), the score is set to 0, and if the value of the aggregated data is less than "m4 - σ4" or more than "m4 - 0.5 x σ4", the score is set to 0. (z value is -1.0 or more and less than -0.5), score = 20, and score is "m4 - 0.5 x σ4" or more and less than "m4" (z value is -0.5 or more and less than 0). If the score is "m4" or more and less than "m4 + 0.5 x σ4" (z value is 0 or more and less than + 0.5) then the score = 60, and if the value is "m4 + 0.5 x σ4" or more and less than "m4 + σ4" (The z value is +0.5 or more and less than +1), the score is set as 80, and the case that the z value is "m4+σ4" or more (the z value is +1.0 or more), the score is set as 100.

With this aspect, each aggregated data in column (B) can be normalized to a range of 0 to 100 points depending on the degree of rarity.

As with the primary office data, these scores indicate the strength of the upward or downward trend in office rental market conditions. For example, a period in which the score of the 4th office data is 100 points means that the aggregated data (building rental industry index) for that period is statistically unusual and large by one standard deviation or more. This suggests that the strength of the rise in office rental market conditions is that strong. On the other hand, a period in which the score of the fourth office data is 0 means that the aggregated data for that period is statistically rare and small by more than one standard deviation, so the office rental market declines. It suggests that the strength to move towards is that much stronger.

Column (I) is a summary of column (H) in one column.
<5th office data management table>

Next, FIG. 7 shows the fifth office data management table 650 stored in the storage device 240. The fifth office data management table 650 is a table that stores fifth office data (also referred to as fifth market data), which is one of the six market data used to calculate Office-RISM.

The fifth office data consists of, for example, aggregate data compiled every three months of the contracted rent DI in Tokyo's 23 wards, including the five central wards, published by XYMAX Real Estate Research Institute, Inc. The contracted rent DI is an index that expresses the change in the contracted rent unit price of a rental office building.For each office building, the contracted rent unit price is compared with the contracted rent unit price in the same building in the previous year, and the building that has increased is determined. minus the percentage of buildings that have fallen.

The contracted rent DI is market data regarding the contract details of rental offices, and its value tends to increase as the office rental market becomes more favorable. Therefore, the contracted rent DI tends to increase or decrease in line with trends in the office rental market.

Column (A) lists the last day of each three-month period. In the example shown in Figure 7, the earliest period is the fourth quarter of 2001 (the last day is December 31, 2001), and the latest period is the first quarter of 2022 (the last day is March 31, 2022). It is.

Column (B) lists aggregated data of the contracted rent DI for every three months.
Column (C) lists the value (m5 - σ5) obtained by subtracting 1.0 x standard deviation (σ5) from the average value (m5) of the aggregated data listed in column (B) (z value = -1). In the example shown in FIG. 7, it is "-32.3". Therefore, if the aggregated data listed in column (B) is smaller than "-32.3", the aggregated data statistically exceeds one standard deviation (σ5) from the average value (m5). You can see that it is small. Note that, as described in columns (J) and (K) of FIG. 7, the average value (m5) of the aggregated data is -0.8, and the standard deviation (σ5) is 31.43.

Column (D) lists the value (m5 - 0.5 x σ5) obtained by subtracting 0.5 x standard deviation from the average value of the aggregated data listed in column (B) (z value = - (equivalent to 0.5).

Column (E) lists the average value (m5) of the aggregated data listed in column (B) (corresponding to z value = 0).

In column (F), the value (m5 + 0.5 x σ5), which is the average value of the aggregated data listed in column (B) plus 0.5 x standard deviation, is listed (z value = 0.5 ).

In column (G), a value (m5+σ5) obtained by adding 1.0×standard deviation to the average value of the aggregated data described in column (B) is written (corresponding to z value = 1).

In Column (H), the score obtained by normalizing the aggregated data listed in Column (B) according to the degree of rarity within the relevant market data (within the 5th office data) is entered. ing.

In this embodiment, if the value of the aggregated data is less than "m5 - σ5" (z value is less than -1.0), the score is set to 0, and if the value of the aggregated data is less than "m5 - σ5" or more than "m5 - 0.5 x σ5", the score is set to 0. (If the z value is -1.0 or more and less than -0.5), the score = 20, and if the z value is "m5 - 0.5 x σ5" or more and less than "m5" (the z value is -0.5 or more and less than 0), the score = 20. If the score is "m5" or more and less than "m5 + 0.5 x σ5" (z value is 0 or more and less than + 0.5), the score = 60, and if the value is "m5 + 0.5 x σ5" or more and less than "m5 + σ5" (The z value is +0.5 or more and less than +1), the score is set as 80, and when the z value is "m5+σ5" or more (the z value is +1.0 or more), the score is set as 100.

With this aspect, each aggregated data in column (B) can be normalized to a range of 0 to 100 points depending on the degree of rarity.

As with the primary office data, these scores indicate the strength of the upward or downward trend in office rental market conditions. For example, a period in which the 5th office data score is 100 points means that the aggregated data (contracted rent DI) for that period is statistically a large value with a rarity of one standard deviation or more. This suggests that the strength of the rise in rental market conditions is that strong. On the other hand, a period in which the score of the 5th office data is 0 means that the aggregated data for that period is statistically rare and small by more than one standard deviation, so the office rental market declines. It suggests that the strength to move towards is that much stronger.

Column (I) is a summary of column (H) in one column.
<6th office data management table>

Next, FIG. 8 shows the sixth office data management table 660 stored in the storage device 240. The sixth office data management table 660 is a table that stores sixth office data (also referred to as sixth market data), which is one of the six market data used to calculate Office-RISM.

The 6th office data consists of, for example, aggregate data compiled every three months of the free rent index in Tokyo's 23 wards, including the five central wards, published by Xymax Real Estate Research Institute, Inc. The free rent index is the average value of the free rent period granted to tenants at the time of contract for a rental office building, normalized to 100 for fiscal year 2003.

The free rent index is market data regarding the details of rental office contracts, and its value tends to decrease as the office rental market becomes more favorable. For this reason, the free rent index tends to increase or decrease in a manner that goes against the trends in the office rental market.

Column (A) lists the last day of each three-month period. In the example shown in Figure 8, the earliest period is the first quarter of 2002 (the last day is March 29, 2002), and the latest period is the first quarter of 2022 (the last day is March 31, 2022). It is.

In column (B), each aggregated data of the free rent index is listed every three months.

In column (C), values obtained by replacing each aggregated data listed in column (B) with a predetermined number (two in this embodiment) of backward moving average values are listed. In other words, each aggregated data in column (C) is a 6-month backward moving average of the aggregated data in column (B). With this aspect, it is possible to reduce variations in the value of the free rent index and obtain aggregated data that more accurately represents subtle changes in trends in the free rent index.

Column (D) lists values obtained by replacing each aggregated data listed in column (B) with the amount of change from the aggregated data a certain period of time ago (four quarters ago in this embodiment).

Column (E) lists values obtained by replacing each aggregated data listed in column (C) with the amount of change from the aggregated data a certain period of time ago (four quarters ago in this embodiment).

Furthermore, when comparing the aggregated data listed in column (D) and the aggregated data listed in column (E), both of them are compared to the aggregated data (free rent index) listed in column (B). They are common in that they are aggregated data generated by performing preprocessing. However, while the former directly replaces the free rent index in column (B) with the amount of change from the previous quarter, the latter uses the free rent index in column (B) once as a 6-month backward moving average. This is converted to reduce the dispersion in the value of the free rent index, and then replaced with the amount of change from the previous quarter. Therefore, the aggregated data listed in column (E) is considered to accurately capture even slight changes in the free rent index.

Column (F) lists the value (m6 - σ6) obtained by subtracting 1.0 x standard deviation (σ6) from the average value (m6) of the aggregated data listed in column (D) (z value = -1). In the example shown in FIG. 8, it is "-0.8". Therefore, if the aggregated data written in column (D) or column (E) is smaller than "-0.8", the aggregated data is statistically one standard deviation away from the average value (m6). It can be seen that it is smaller than (σ6). Note that, as described in columns (M) and (N) of FIG. 8, the average value (m6) of the aggregated data is 0.064, and the standard deviation (σ6) is 0.87.

Column (G) lists the value (m6 - 0.5 x σ6) obtained by subtracting 0.5 x standard deviation from the average value of the aggregated data listed in column (D) (z value = - (equivalent to 0.5).

In column (H), the average value (m6) of the aggregated data written in column (D) is written (corresponding to z value = 0).

In Column (I), the value (m6 + 0.5 x σ6), which is the average value of the aggregated data listed in Column (D) plus 0.5 x standard deviation, is listed (z value = 0.5 ).

In column (J), a value (m6+σ6) obtained by adding 1.0× standard deviation to the average value of the aggregated data written in column (D) is written (corresponding to z value = 1).

Column (K) contains the score obtained by normalizing the aggregated data listed in column (E) according to the degree of rarity within the relevant market data (within the 6th office data). ing.

In this embodiment, if the value of the aggregated data is less than "m6 - σ6" (z value is less than -1.0), the score = 100, and if the value is "m6 - σ6" or more and less than "m6 - 0.5 x σ6", the score is set as 100. (z value is -1.0 or more and less than -0.5), score = 80, and score is "m6 - 0.5 x σ6" or more and less than "m6" (z value is -0.5 or more and less than 0). If the score is "m6" or more and less than "m6 + 0.5 x σ6" (z value is 0 or more and less than + 0.5), the score = 40, and if the value is "m6 + 0.5 x σ6" or more and less than "m6 + σ6" (The z value is +0.5 or more and less than +1), the score is set as 20, and the case that the z value is "m6+σ6" or more (the z value is +1.0 or more), the score is set as 0.

With this aspect, each aggregated data in column (E) can be normalized to a range of 0 to 100 points depending on the degree of rarity.

As with the primary office data, these scores indicate the strength of the upward or downward trend in office rental market conditions. For example, a period in which the score of the 6th office data is 100 points means that the aggregated data (free rent index) for that period is statistically rare and small by one standard deviation or more. This suggests that the strength of the rise in rental market conditions is that strong. On the other hand, a period in which the No. 6 office data score is 0 points means that the aggregated data for that period is statistically unusual and large by one standard deviation or more, so the office rental market declines. It suggests that the strength to move towards is that much stronger.

Column (L) is a compilation of column (K) in one column.
<Leading indicator management table>

Next, FIG. 9 shows a leading indicator management table 670 stored in the storage device 240. The leading indicator management table 670 is a table that stores leading indicators (Office-RISM) calculated using the first to sixth office data described above.

Columns (A) to (F) are for each three-month period from the first quarter of 2000 (the last day is March 31, 2000) to the first quarter of 2022 (the last day is March 31, 2022). Each score of the first office data to the sixth office data calculated in each period is stored.

In column (G), the number of scores calculated in each period is described. For example, if all the scores from the first office data to the sixth office data are the same, the score is 6.

In column (H), representative values for each period of scores for each office data are listed. Although an average value, maximum value, minimum value, median value, mode value, etc. can be used as the representative value for each period, the average value is used in this embodiment. The representative value for each of these periods is the leading indicator (Office-RISM). FIG. 10 shows how these representative values are displayed on a graph in time series.

As mentioned above, Office-RISM is a leading indicator of the office rental market, and as shown in Figure 21, the average recruitment published by Miki Shoji Co., Ltd., which is one of the benchmarks for the office rental market, is a leading indicator of the office rental market. It is almost exactly several years ahead of changes in rents. In this way, Office-RISM makes it possible to predict future trends in the office rental market.

Although details will be described later, as shown in Figure 22, Office-RISM can also be used to determine the phase of the office rental market fluctuation cycle (recovery phase, boom phase, recession phase, recession phase). For this reason, Office-RISM is not only useful for predicting future trends in the office rental market, but also a useful indicator for determining the current state of the office rental market.

<Functional configuration of leading indicator calculation device>
Next, the functional configuration of the leading indicator calculation device 200 will be described with reference to the functional configuration diagram shown in FIG. 20.

As described above, the leading indicator calculation device 200 operates by reading the leading indicator calculation device control program 700 and various data stored in the storage device 240 into the memory 220 and executing or processing them by the CPU 210. Various functions as the index calculation device 200 are realized.

Specifically, the leading indicator calculation device 200 has the following functions: a market condition data acquisition section 201 , a preprocessing execution section 202 , a leading indicator calculation section 203 , and a market condition determination section 204 .

The market data acquisition unit 201 acquires aggregate data for each predetermined period (every three months in this embodiment) of multiple types (six types in this embodiment) of market data that change according to trends in the office rental market. . The six types of market data are the first office data to the sixth office data, as described above.

Incidentally, in order to calculate a more accurate and more advanced leading indicator of office rental market conditions, it is important to decide which market data to use from among a large number of market data. The six types of office data according to this embodiment are selected through a process as shown in FIG.

First, from a large amount of market data, we narrowed it down to real estate-related market data by excluding so-called macroeconomic indicators that are not limited to the real estate industry, such as GDP and unemployment rate. Based on this, we selected approximately 50 types of market data related to office rental market conditions. In this way, by excluding macroeconomic indicators and specializing in market data related to office rental market conditions, it is possible to identify trends in the office rental market, such as the decline in demand for offices due to the spread of telework in recent years. Even in cases where trends do not match economic trends, it is now possible to accurately grasp trends in the office rental market.

Next, we identify the point in time when the average asking rent published by Miki Shoji Co., Ltd., which is one of the benchmarks for the office rental market, changes from positive to negative or from negative to positive compared to the previous year or the previous quarter. We selected market data that showed similar movements prior to that point in time.

Then, from among these market data, we selected six types of market data that show particularly accurate trends in the office rental market. This is the first office data to the sixth office data. At that time, these market data include multiple types of market data, such as market data on the operating status of rental offices, market data on the contract details of rental offices, and market data on business sentiment held by rental office managers. We have included market data that explains office rental market conditions from different perspectives. This approach makes it possible to comprehensively understand trends in the office rental market from multiple perspectives. For example, even if some special factors cause exceptional fluctuations in some market data, the impact can be minimized. It becomes possible to do so.

Returning to FIG. 20, the preprocessing execution unit 202 performs predetermined preprocessing on this market data so that at least some of the market data changes in advance of changes in the office rental market. .

For example, the preprocessing execution unit 202 replaces each aggregated data of at least some of the market condition data with the amount of change from the aggregated data a certain period of time ago. As described above, in this embodiment, such preprocessing is performed on the first office data (percentage of fully occupied buildings), the third office data (vacancy rate), and the sixth office data (free rent index). is being implemented.

Due to this aspect, each aggregated data after preprocessing represents the strength of change when each aggregated data before preprocessing moves toward a peak (maximum value) or bottom (minimum value). Therefore, it can be said that the data has been converted into data that is more advanced than the market data before preprocessing.

Further, the preprocessing execution unit 202 replaces, for example, each aggregated data of at least some of the market condition data with a predetermined number of backward moving average values. As described above, in this embodiment, such a priori is applied to the second office data (net absorption), the fourth office data (building management business sentiment), and the sixth office data (free rent index). Processing is being carried out.

This mode reduces the variation in the aggregated data for each period, and makes it possible to detect subtle changes in market data that would otherwise be buried among the variations in the aggregated data before preprocessing. , it becomes possible to detect changes in trends in market data at an earlier stage.

However, although preprocessing using a backward moving average can alleviate variations in each piece of aggregated data, it also has the effect of delaying changes in trends in market data because each piece of aggregated data is dragged by past values. Therefore, when performing preprocessing using a backward moving average, it is important to take these influences into consideration and determine the optimal number of data items (period) for performing the moving average.

In this embodiment, as shown in FIGS. 12 to 19, it is verified what kind of preprocessing should be performed on each of the first to sixth office data.

First, FIG. 12 shows a summary of the verification results performed on the first to sixth office data.

In this embodiment, regarding the backward moving average, "no time difference", "2 quarter moving average", "3 quarter moving average" (omitted in FIG. 12), and "4 quarter moving average" were verified.

"No time difference" indicates that backward moving average preprocessing is not performed. "Two-quarter moving average" indicates that each aggregated data is replaced with the average value of the aggregated data from three months ago (the average value of two consecutive pieces of aggregated data). "3-quarter moving average" indicates that each aggregated data is replaced with the average value of aggregated data 6 months ago and 3 months ago (average value of three consecutive pieces of aggregated data). "4-quarter moving average" indicates that each aggregated data is replaced with the average value of the aggregated data from 9 months ago, 6 months ago, and 3 months ago (average value of 4 consecutive pieces of aggregated data) .

In addition, regarding the preprocessing of replacing each aggregated data with the amount of change from the aggregated data a certain period of time ago, we verified the ``original figures,'' ``previous period difference (ratio),'' and ``previous year difference (ratio).''

"Original value" indicates that no preprocessing is performed to replace the amount of change. "Previous period difference (ratio)" indicates that each aggregate data is replaced with the amount of change (or rate of change) from the aggregate data three months ago (quarter ago). “Year-on-year difference (ratio)” indicates that each aggregated data is replaced with the amount of change (or rate of change) from the aggregated data 12 months ago (four quarters ago).

In the table shown in FIG. 12, the column labeled "t" indicates that the degree of delay of each aggregated data after preprocessing exceeds the allowable range and is difficult to adopt as preprocessing. As can be seen from FIG. 12, the longer the period for which backward moving average is performed, the more lagging the data becomes, and the stronger the tendency for all market data to be classified as "t".

Note that the criteria for the above-mentioned allowable range may or may not be determined as a specific numerical range. In the latter case, it is sufficient to judge whether or not the data can be adopted by making a comprehensive judgment while comparing the quality of each aggregated data after preprocessing with the results of other preprocessing.

On the other hand, the column labeled "m" in FIG. 12 indicates that the degree of variation in each aggregated data after preprocessing exceeds the allowable range, and it is difficult to adopt it as preprocessing. As can be seen from Figure 12, the dispersion of each aggregated data becomes larger as the period in which backward moving averages are applied for any market condition data becomes shorter; ” and “Free Rent Index,” this tendency is particularly strong.

From the above, among the six market conditions data, backward moving averages are not performed for three of the ``full occupancy building ratio'', ``vacancy rate'', and ``contracted rent DI'', and ``net absorption'' and ``building management We decided to perform preprocessing using backward moving averages for three items: ``Business Confidence'' and ``Free Rent Index.'' The situation is schematically shown in FIG.

Next, with reference to FIGS. 14 to 19, a process for examining the content of preprocessing to be applied to the six market data will be described.

First, regarding the "full occupancy building ratio" (first office data), we tried various preprocessing methods shown in the table in Figure 12 for each aggregated data, and found that the backward moving average preprocessing has a tendency to delay the data. It was difficult to tolerate. Therefore, preprocessing (Case A) that replaces each aggregated data with the amount of change from the aggregated data one period ago (quarter ago), and preprocessing (Case A) that replaces each aggregated data with the amount of change from the aggregated data four periods ago (same period last year) B) remained as a candidate.

Then, as shown in Figure 14, when we look at each aggregated data after preprocessing, we see that in case A, there is a large variation in each aggregated data, which is difficult to accept in terms of capturing trends in the office rental market. Ta. On the other hand, in case B, the variation in each aggregated data was suppressed and was judged to be acceptable.

Therefore, for the ``percentage of fully occupied buildings,'' we decided to adopt preprocessing (Case B) in which each aggregated data is replaced with the amount of change from the aggregated data four periods ago.

Regarding "Net Absorption" (second office data), various pre-processing shown in the table of FIG. 12 was similarly tried on each aggregated data. As a result, in pursuit of speed of prediction (precedence), there are case A in which no preprocessing is performed, and case B in which preprocessing is performed in which each aggregated data is replaced with the average value of the aggregated data from one period before. Case C, which performs preprocessing to replace each aggregated data with the amount of change from the aggregated data one period before, remained as a candidate.

Based on that, when we look at the aggregated data for each case as shown in Figure 15, we find that in both cases A and C, there is a large variation in each aggregated data, which is acceptable in terms of capturing trends in the office rental market. It was difficult. On the other hand, in case B, each aggregated data was appropriately leveled and was confirmed to be effective as data for capturing trends in the office rental market.

Therefore, for "net absorption," we decided to adopt preprocessing (case B) in which each aggregated data is replaced with the average value of the aggregated data from the previous period.

Regarding the "vacancy rate" (third office data), various pre-processing shown in the table of FIG. 12 was similarly tried on each aggregated data. As a result, it was difficult to tolerate the data delay in the backward moving average preprocessing. Therefore, preprocessing (case A) in which each aggregated data is replaced with the amount of change from the aggregated data one period ago, and preprocessing (case B) in which each aggregated data is replaced with the amount of change from the aggregated data four periods earlier (the same period last year). remained as a candidate.

Then, when looking at the aggregated data for each case as shown in FIG. 16, it was confirmed that in case A, it was difficult to read the trend of change, especially around 0%. On the other hand, in case B, it was confirmed that the trend of change was clearly understood.

Therefore, for the "vacancy rate," we decided to adopt preprocessing (Case B) in which the amount of change from the aggregated data from four quarters ago (the same period last year) was replaced.

Similarly, various pre-processing steps shown in the table of FIG. 12 were tried for each aggregated data regarding "Building Manager Business Sentiment" (Fourth Office Data). As a result, it was difficult to use differences from the previous quarter or the previous year because it was difficult to grasp the changing trends. Therefore, Case A, in which no preprocessing is performed, and Case B, in which preprocessing is performed to replace each aggregated data with the average value of the aggregated data from one period before, remained as candidates.

Then, when looking at the aggregated data for each case as shown in FIG. 17, it was confirmed that in case A, there was a period in which the aggregated data had large variations. On the other hand, in case B, it was confirmed that the aggregated data was moderately leveled and that the trend of change could be clearly understood.

Therefore, for "Building Managers' Business Sentiment," we decided to adopt preprocessing (Case B) in which each aggregated data is replaced with the average value of the aggregated data from one period before.

Similarly, for the "contracted rent DI" (fifth office data), various pre-processings shown in the table of FIG. 12 were tried on each aggregated data. As a result, it was difficult to use the difference from the previous period or the year before because it was difficult to grasp the trend of change, and the backward moving average preprocessing was difficult to tolerate because of the delay in data. On the other hand, as shown in FIG. 18, it was confirmed that the tendency of change could be grasped even without pretreatment.

Therefore, we have decided not to perform any pre-processing on the "contracted rent DI."

Regarding the "Free Rent Index" (Sixth Office Data), various pre-processing steps shown in the table of FIG. 12 were similarly tried on each aggregated data. As a result, it was possible to understand the trend of change even after preprocessing the differences from the previous period and the year before. On the other hand, the backward moving average of three quarters or more had data lag that was difficult to tolerate. Therefore, preprocessing (case A) that replaces each aggregated data with the amount of change from the aggregated data one period ago without using a backward moving average, and replacing each aggregated data with the amount of change from the aggregated data of the previous year (same period of the previous year) without using a backward moving average. The remaining candidates were preprocessing (Case B) that replaced the amount of change from aggregated data, Case C that combined Case A with a two-quarter backward moving average, and Case D that combined Case B with a two-quarter backward moving average.

Then, when looking at the aggregated data for each case as shown in Figure 19, it was confirmed that in Case A and Case C, there were many periods where the trend of change could not be discerned at the rent turning point. Furthermore, for Case B, the results show that some data have large variations. On the other hand, in case D, it was confirmed that the aggregated data was moderately leveled, and the trend of change could be clearly understood.

Therefore, for the "free rent index," we decided to adopt preprocessing (case D) in which each aggregated data is smoothed by performing a two-period moving average, and then replaced with the amount of change from the aggregated data four periods earlier.

Returning to FIG. 20, the leading indicator calculation unit 203 converts the aggregated data for each period for each of the six types of market data into a score according to the degree of rarity within the market data, and The representative value for each period determined from the score for each period of market data is calculated as a leading indicator (Office-RISM).

Specifically, as explained with reference to FIGS. 3 to 9, the aggregated data of each office data is first divided into 0, 20, 40, 60, 80, or 100 depending on the z value of each aggregated data. (Figures 3 to 8). The leading indicator (Office-RISM) is then calculated by averaging these scores every three months (Figure 9).

Such an aspect makes it possible to calculate a more accurate and more advanced indicator of office rental market conditions. In other words, in the present embodiment, preprocessing is performed on at least some of the multiple types of market data that change according to trends in the office rental market so that the changes start earlier. Above, the aggregated data for each period of each market data is normalized by scoring it out of 100 according to the degree of rarity within the market data, and by averaging these scores for each period. Leading indicators are being calculated. As a result, the leading indicator (Office-RISM) calculated by this embodiment is accurate because it is derived from multiple types of market data that have a high correlation with office rental market conditions, and because it is obtained by performing appropriate preprocessing. It has the characteristics of promoting precedence.

The market condition determining unit 204 determines a change in the office rental market situation based on the above-mentioned leading indicators.

For example, as shown in FIG. 22, if the Office-RISM score is 50 or higher and the Office-RISM score has increased for two consecutive periods, the market condition determination unit 204 determines that the office rental market has recovered. It is determined that the situation has changed. The graph shown in Figure 21 shows that the average asking rent did indeed bottom out and start rising around 2004-2005 and 2012-2013, confirming that the market has entered a recovery phase.

In addition, the market condition determination unit 204 determines that the office rental market has switched to a booming phase if the Office-RISM score is 50 or higher and the Office-RISM score has decreased for two consecutive periods. In the graph shown in Figure 21, the average asking rent did indeed continue to rise around 2006 to 2007 and around 2014 to 2016, confirming that the economy had entered a booming phase.

In addition, the market condition determination unit 204 determines that the office rental market has entered a recession phase if the Office-RISM score is less than 50 and the Office-RISM score has decreased for two consecutive periods. In the graph shown in Figure 21, the average asking rent has certainly turned from increasing to decreasing around 2001-2002, 2008-2009, and 2020, confirming that we have entered a recessionary phase. .

Further, the market condition determination unit 204 determines that the office rental market has entered a recession phase if the Office-RISM score is less than 50 and the Office-RISM score has increased for two consecutive periods. In the graph shown in Figure 21, the average asking rent has certainly continued to decline around 2003 and from 2010 to 2011, confirming that the country has entered a recessionary phase.

In this way, by using the leading indicator (Office-RISM) according to the present embodiment, it is possible to accurately determine the change in the state of the office rental market.

In the example shown in Figure 22, the change in the office rental market situation is determined depending on whether the score of the leading indicator (Office-RISM) is 50 or higher and whether it has increased (or decreased) for two consecutive periods. Although examples of determination have been shown, the numerical values used as the criteria for these determinations are merely examples, and other criteria for determination may be adopted. For example, if the type or number of market data on which the leading indicators are based differs, a different criterion for determination may be adopted.

Therefore, to explain the function of the market condition determination unit 204 according to the present embodiment more generally, the representative value of the determination target period among the representative values of the scores of each period forming the leading indicator is equal to or higher than a predetermined value (for example, 50 or higher). , and if a predetermined number (for example, 2) of consecutive representative values in the past including the representative value of the determination period are on an increasing trend, it is determined that the office rental market situation in the determination period has switched to a recovery phase. , if the representative value for the determination period is greater than or equal to a predetermined value, and a predetermined number of representative values are on a decreasing trend, it is determined that the office rental market for the determination period has entered a boom phase, and the representative value for the determination period is is less than a predetermined value, and the representative value of the predetermined number of items is in a decreasing trend, it is determined that the office rental market situation during the determination period has entered a recession phase, and the representative value during the determination period is less than the predetermined value, If the predetermined number of representative values is on an increasing trend, it is determined that the office rental market conditions during the determination period have entered a recession phase.

Note that, after switching to one of the above situations, the market condition determination unit 204 maintains the original situation until switching to another situation. In addition, the period to be determined may be any one of the periods that arrive every predetermined period (every three months) (in this embodiment, the period from the first quarter of 2000 to 2022, when the leading indicators are calculated). any period during the first quarter). Regarding the four phases of the office rental market, please refer to A. F. Burns and W. C. It is based on the definition by Mitchell.

==Processing flow==
Next, the flow of processing by the leading indicator calculation device 200 according to this embodiment will be described with reference to the flowchart shown in FIG. 25.

First, the leading indicator calculation device 200 obtains aggregated data for each predetermined period (every three months in this embodiment) of multiple types (six types in this embodiment) of market data that change according to trends in the office rental market. (S1000). Further, at this time, the leading indicator calculation device 200 performs predetermined preprocessing on at least part of the market condition data so that the change occurs more in advance of changes in the office rental market condition.

Next, for each market data, the leading indicator calculation device 200 converts the aggregated data for each period into a score according to the degree of rarity within the market data (S1010 to S1030). Specifically, the leading indicator calculation device 200 calculates the average value and standard deviation of the aggregated data for each market data (S1010), and also indicates how many times the standard deviation each aggregated data is away from the average value. The index (for example, z value) is associated with the score (S1020). In this embodiment, the above-mentioned index is "less than -1", "-1 or more and less than -0.5", "-0.5 or more and less than 0", "0 or more and less than 0.5", and " If the score is in the range of 0.5 or more and less than 1, and 1 or more, then the score is set as 0, 20, 40, 60, 80, or 100, respectively. , "100", "80", "60", "40", "20", or "0". The leading indicator calculation device 200 then converts each aggregated data into a score according to the z value of each aggregated data (S1030). Of course, the above correspondence between the z value and the score is just an example, and other correspondences may be used.

The leading indicator calculation device 200 calculates the representative value (in this embodiment, the average value) for each period of the scores of these multiple types of market condition data as a leading indicator (Office-RISM) (S1040).

Thereafter, the leading indicator calculation device 200 can perform various analyzes of the office rental market using this leading indicator (S1050 to S1070).

For example, the leading indicator calculation device 200 obtains the aggregated data of the average asking rent published by Miki Shoji Co., Ltd. for each predetermined period (for example, every 3 months) as a benchmark for office rental market conditions (S1050), and calculates the average asking rent. Comparative analysis of asking rents and Office-RISM, judgment of the current state of the office rental market, and understanding of turning points in office rents are performed (S1060).

Figure 21 shows the results of a comparative analysis between the average asking rent and Office-RISM. Looking at Figure 21, for example, when Office-RISM becomes 50 or higher (increase signal lit) and then reaches around 80, the average asking rent starts to rise, and when it becomes less than 50 (decrease signal lit) and then reaches around 20, the average asking rent starts to rise. We can confirm that the average asking rent is starting to decline.

Furthermore, as of 2022, the average asking rent has entered a downward trend after peaking in 2020, indicating that the office rental market is in a recessionary phase, gradually approaching a recession. However, Office-RISM has bottomed out around 2021 and is starting to rise, so although the average asking rent will fall for the time being, it can be expected that the situation will change towards an upward trend. In the past, it took about two to three years from the reversal of Office-RISM until the average asking rent rose, and during that time the average asking rent declined by about 10%. It can be predicted that rents will fall by about 10%, but will then trend upward.

Furthermore, Figure 23 shows the results of an analysis for predicting the extent to which the average asking rent will fall in the declining phase that started in 2020. In the past, average asking rents have fallen between 2002 and 2005 and between 2008 and 2013, but the decline was greater in the latter period. Therefore, when we tallied up the number of office data with a score of "0" during these periods, we found that in the period from 2002 to 2005, there were at most three office data (that is, three office data had a score of "0"). However, in the period from 2008 to 2013, there were a maximum of 6 (all office data scores were "0"). From this, it is expected that the greater the number of office data with a score of "0" during the same period, the greater the decline or rate of decline in average asking rent. Applying this analysis result to the declining phase of average asking rent from 2020 onward, we find that the number of office data with a score of "0" during the same period is 4, so It is predicted that the decline will not be that large.

Further, the leading indicator calculation device 200 can also verify (cumulative analysis) the stability of multiple types (six types) of office data employed to calculate Office-RISM (S1070). In the example shown in Figure 24, the results obtained by subtracting 50 from each Office-RISM score and dividing by 5 can be roughly traced to the average asking rent. The stability of each office data can be confirmed.

The leading indicator calculation device 200, the control method for the leading indicator calculation device 200, and the leading indicator calculation device control program 700 according to the present embodiment have been described above in detail. It becomes possible to calculate a leading indicator (Office-RISM) of a certain office rental market situation. As shown in Figure 21, for example, this leading indicator can accurately suggest the turning points of increases and decreases in average asking rents over the period from 2000 to 2022.

Also, as can be seen from Figure 21, Office-RISM indicates that when it exceeds 50, it suggests a shift to an upward trend in the average asking rent, and when it falls below 50, it suggests a shift to a downward trend in the average asking rent. It is also possible to detect trend turning points in the office rental market. For this reason, it can be said to be extremely useful, for example, in determining the timing of acquiring or selling an office building, or negotiating the rent of an office building.

Additionally, Office-RISM is created using multiple types of publicly available market data, so it is objective, quantitative, and highly reliable. Since these market data do not include macroeconomic indicators that are not limited to the real estate industry, for example, the impact of the coronavirus pandemic on the office rental market is also factored in, and the office rental data also takes into account structural changes. It is now possible to predict market conditions.

Office-RISM also has simple and easy-to-understand calculation procedures, making it easy to replace, add, or delete the underlying market data, or change preprocessing, making it highly versatile.

By using these leading indicators of the office rental market, which have many excellent features, it is possible to comprehensively understand the ups and downs and direction of the office rental market.

Note that the embodiments described above are for facilitating understanding of the present invention, and are not intended to be interpreted as limiting the present invention. The present invention may be modified and improved without departing from its spirit, and the present invention also includes equivalents thereof.

For example, in this embodiment, the leading indicator (Office-RISM) is calculated using six market condition data from the first office data to the sixth office data, but the number of office data used to calculate the leading indicator is six. The number is not limited to , and it is sufficient if there are two or more.

Furthermore, the number of phases of the office rental market that the market condition determination unit 204 determines based on leading indicators is not limited to four, but may be two (boom phase and recession phase), or three or more (for example, deceleration, stagnation). , adjustment, deterioration, bottoming out, pick-up, recovery, and expansion).

200 Leading indicator calculation device 201 Market condition data acquisition unit 202 Preprocessing execution unit 203 Leading indicator calculation unit 204 Market condition determination unit 210 CPU
220 Memory 230 Communication device 240 Storage device 250 Input device 260 Output device 270 Recording medium reading device 500 Network 610 First office data management table 620 Second office data management table 630 Third office data management table 640 Fourth office data management table 650 Fifth office data management table 660 Sixth office data management table 670 Leading indicator management table 700 Leading indicator calculation device control program 800 Recording medium

Claims (11)

A leading indicator calculation device for calculating leading indicators of office rental market conditions,
a market data acquisition unit that acquires aggregated data for each predetermined period of multiple types of market data that change according to trends in the office rental market;
before performing predetermined preprocessing on at least some of the market condition data so that the at least some of the market condition data changes in advance of changes in office rental market conditions or the variation is alleviated; a processing execution unit;
For each of the market data, aggregate data for each period is converted into a score according to the degree of rarity within the market data, and a representative value for each period is determined from the scores for each period of the plurality of types of market data. a leading indicator calculation unit that calculates the leading indicator;
A leading indicator calculation device comprising: The leading indicator calculation device according to claim 1,
The multiple types of market data include:
Contains market data regarding the occupancy status of rental offices, market data regarding contract details of rental offices, and market data regarding business sentiment held by rental office managers.
Leading indicator calculation device. The leading indicator calculation device according to claim 1,
The preprocessing execution unit executes the preprocessing by replacing each aggregated data of at least some of the market condition data with an amount of change from the aggregated data a certain period ago, respectively.
Leading indicator calculation device. The leading indicator calculation device according to claim 1,
The preprocessing execution unit executes the preprocessing by replacing each aggregated data of at least some of the market data with a predetermined number of backward moving average values, respectively.
Leading indicator calculation device. The leading indicator calculation device according to claim 1,
The leading indicator calculation unit includes:
For market data in which the value of the aggregated data tends to increase as the office rental market becomes more favorable,
If the z value of the aggregated data is less than -1.0, the score of the aggregated data is set to 0,
If the z value of the aggregated data is -1.0 or more and less than -0.5, the score of the aggregated data is set to 20,
If the z value of the aggregated data is -0.5 or more and less than 0, the score of the aggregated data is set to 40,
If the z value of the aggregated data is 0 or more and less than 0.5, the score of the aggregated data is set to 60,
If the z value of the aggregated data is 0.5 or more and less than 1.0, the score of the aggregated data is set to 80,
If the z value of the aggregated data is 1.0 or more, the score of the aggregated data is set to 100,
For market data in which the value of the aggregated data tends to decrease as the office rental market becomes more favorable,
If the z value of the aggregated data is less than -1.0, the score of the aggregated data is set to 100,
If the z value of the aggregated data is -1.0 or more and less than -0.5, the score of the aggregated data is set to 80,
If the z value of the aggregated data is -0.5 or more and less than 0, the score of the aggregated data is set to 60,
If the z value of the aggregated data is 0 or more and less than 0.5, the score of the aggregated data is set to 40,
If the z value of the aggregated data is 0.5 or more and less than 1.0, the score of the aggregated data is set to 20,
If the z value of the aggregated data is 1.0 or more, the score of the aggregated data is set to 0;
Leading indicator calculation device. The leading indicator calculation device according to claim 1,
A leading indicator calculation device further comprising a market condition determination unit that determines a change in the phase of the office rental market based on the leading indicator. The leading indicator calculation device according to claim 6,
The market condition judgment department is
Among the representative values of each period forming the leading indicator, the representative value of the judgment target period is equal to or greater than a predetermined value, and a predetermined number of consecutive representative values in the past including the representative value of the judgment target period are on the increase. In certain cases, it is determined that the office rental market has entered a recovery phase during the determination period;
If the representative value of the determination target period is equal to or higher than a predetermined value, and the representative value of the predetermined number of items is on a decreasing trend, it is determined that the office rental market has switched to a booming phase in the determination target period;
If the representative value of the determination target period is less than a predetermined value and the representative value of the predetermined number of items is in a decreasing trend, determining that the office rental market has entered a recession phase in the determination target period;
If the representative value for the determination period is less than a predetermined value and the predetermined number of representative values is on an increasing trend, determining that the office rental market has entered a recessionary phase in the determination period;
Leading indicator calculation device. The leading indicator calculation device according to claim 1,
The multiple types of market data are
first market condition data consisting of aggregated data of the percentage of rental office buildings operating with full occupancy for each predetermined period;
second market condition data consisting of aggregated data of the net absorption of rental office buildings for each of the predetermined periods;
third market condition data consisting of aggregated data of vacancy rates of rental office buildings for each of the predetermined periods;
Fourth market condition data consisting of aggregated data obtained by aggregating a real estate industry business index indicating the business outlook for the next three months in the building rental industry for each of the predetermined periods;
fifth market condition data consisting of aggregated data obtained by aggregating the contracted rent DI representing changes in the contracted rent unit price of the rental office building for each predetermined period;
Sixth market condition data consisting of aggregate data obtained by aggregating the average value of free rent periods granted to tenants of rental office buildings for each predetermined period;
A leading indicator calculation device including at least two of the following. The leading indicator calculation device according to claim 1,
The leading indicator calculating unit is a leading indicator calculating device that calculates an average value for each period of scores of the plurality of types of market condition data as a representative value for each period. A method for controlling a leading indicator calculation device for calculating a leading indicator of office rental market conditions, the method comprising:
The leading indicator calculation device includes:
Obtain aggregated data for each predetermined period of multiple types of market data that change according to trends in the office rental market,
performing predetermined preprocessing on at least some of the market condition data so that at least some of the market condition data changes in advance of changes in office rental market conditions, or so that variations are alleviated ;
For each of the market data, aggregate data for each period is converted into a score according to the degree of rarity within the market data, and a representative value for each period is determined from the scores for each period of the plurality of types of market data. A method for controlling a leading indicator calculation device, which calculates the leading indicator. A program for causing a computer to calculate leading indicators of office rental market conditions,
to the computer;
A procedure for acquiring aggregated data for each predetermined period of multiple types of market data that changes according to trends in the office rental market;
A step of performing predetermined pre-processing on at least some of the market condition data so that the at least some of the market condition data changes in advance of changes in office rental market conditions or the variation is alleviated. and,
For each of the market data, aggregate data for each period is converted into a score according to the degree of rarity within the market data, and a representative value for each period is determined from the scores for each period of the plurality of types of market data. A procedure for calculating the leading indicator;
A program to run.

Images