TWM624436U - Housing price appraisal equipment - Google Patents

Abstract

A housing price evaluation device includes a storage device, an inquiry device and an evaluation device. The inquiry device is used for receiving an inquiry building signal indicating an area and electrically connected to the storage device, and performing a data look-up table on a database according to the area information of the inquiry building signal to generate an inquiry result signal. The query result signal indicates a pair of building data corresponding to the area information. The estimating device is connected to the query device for receiving the query result signal from the query device, and calculates and generates a house price loss rate according to the building data (including local flooding information) and an estimation formula.

Description

Home Price Appraisal Equipment

The new model relates to a kind of equipment, especially a kind of house price evaluation equipment.

According to the current financial report on climate-related risks disclosed by the financial industry, the current climate scenario analysis is based on the inventory of risk cases or the adjustment of existing credit risk model parameters, and after calculating the difference in the value of the collateral or the expected loss of the case, it is disclosed that the collateral is located in the climate. Risk area exposure amount or expected loss. For example, counting the number of climate risk cases and their proportion to the total financing amount cannot effectively quantify the expected loss amount of each collateral or each risk area, and is still insufficient in the completeness of the situational loss analysis. Or adjust the parameters of the credit risk model (default rate and loss given default rate) of bank credit customers, and calculate the difference in expected loss of credit cases with high climate risk, which cannot reflect the price loss caused by the location and type of the building (collateral).

In addition, when the relevant literature evaluates the impact of flooding disasters on housing prices, most of the analysis methods adopt a single methodology, which cannot analyze areas with insufficient data, or is limited by the difficulty of collecting flooding information, so the analysis area is relatively narrow, and it is not possible to fully analyze the entire Taiwanese. The impact of flooding disasters on housing prices.

Therefore, the purpose of the present invention is to provide a housing price evaluation device that considers the impact of flooding disasters.

Therefore, the present invention includes a storage device, an inquiry device and an evaluation device.

The storage device has a database. The database records a plurality of building data, and the plurality of building data respectively correspond to a plurality of regions.

The query device is used for receiving a query building signal indicating an area and electrically connected to the storage device, and performing a data lookup table on the database according to the query building signal to generate a query result signal. The query result signal indicates a pair of building data corresponding to the area information. The building data includes a piece of flooding information, a number of historical flooded areas related to the flooding information, and a number of historical non-flooded areas, wherein the historical non-flooded area is the edge area of the historical flooded area.

The estimating device is connected to the query device for receiving the query result signal from the query device, and generating a house price loss rate according to the building data and an estimation formula. This estimation formula is related to a sample adequacy. The definition of the sample sufficiency is that the number of pieces of the historical flooded area and the historical non-flooded area related to the corresponding area are respectively greater than or equal to a minimum demand quantity.

The utility model has the following functions: the estimating device generates the house price loss rate according to the building data and the estimating formula. The estimation formula is related to the sample adequacy. The definition of the sample adequacy is that the number of historical flooded areas and historical non-flooded areas related to the corresponding area are respectively greater than or equal to the minimum demand quantity.

Referring to FIG. 1 and FIG. 2 , an embodiment of the novel housing price evaluation device is shown. The house price evaluation device includes a communication interface device 1 , a query device 2 , a storage device 3 and an evaluation device 4 .

In step 501, the communication interface device 1 is electrically connected to the query device 2, and is used to generate a building query signal according to an input command.

The storage device 3 has a database. The database records a plurality of building data, and the plurality of building data respectively correspond to a plurality of regions.

In step 502, the query device 2 is used to receive the query building signal indicating an area and electrically connect to the storage device 3, and perform a data lookup table on the database according to the query building signal to generate a query result signal . The query result signal indicates a pair of building data corresponding to the area information. The building information includes a house feature, a geographic information, a public information, a real price login, a community information, a flooding information and a number of historical flooded areas and historical non-flooded areas related to the flooding information where the historically non-flooded area is the marginal area of the historically flooded area. The house features include house age, floor, administrative area, house area and other features. The geographic information includes variables such as range, distance, and function. The public information includes indicators such as social economy, population, public security, disasters, and pollution. The actual price entry includes transaction prices in different periods and ranges. The community information includes the transaction prices of the same community and the same developer in different periods. The flooding information includes a historical flooding depth and a potential flooding depth corresponding to the area.

In step 503, the estimating device 4 is connected to the querying device 2 to receive the query result signal from the querying device 2, and substitute an estimation formula according to the building data to generate a house price loss rate. The estimation formula is related to a sample adequacy and a modeling significance. The definition of the sample sufficiency is that the number of pieces of the historical flooded area and the historical non-flooded area related to the corresponding area are respectively greater than or equal to a minimum demand quantity. The modeling significance is related to whether a linear regression model can be built.

For example, the minimum required quantity is 30 pieces. Referring to FIG. 3 , in step 601 , the query device 2 queries the database, and the database has samples of all Taiwan (administrative regions/counties and cities) in the row. In step 602 , when the evaluation device 4 determines that the number of the historical flooded area and the historical non-flooded area corresponding to the area is greater than or equal to 30 pieces, respectively, the building data has the sample sufficiency. In step 602 , when the estimating device 4 determines that one of the number of pieces in the historical flooded area and the historical non-flooded area corresponding to the area is less than 30 pieces, the estimating formula is a Bootstrap Method.

，其中，y是應變數，

是迴歸係數，

是該建物資料經過換算後對應的建模代表值，

是誤差值。

是由一多變數迴歸係數公式計算而得。需注意的是，該線性迴歸公式是以一行政區為最小單位進行建模，若行政區樣本未滿足該樣本充足性，則以縣市為最小單位進行建模。以臺北市大安區為例，該建物資料如表1所示。 In step 603, when the evaluation device 4 determines that the building data has linear regression, the building data has the modeling significance. In step 604, the estimation formula is a linear regression formula, and the linear regression formula is used to establish a linear regression model (hereinafter referred to as modeling). The linear regression formula

, where y is the strain number,

is the regression coefficient,

is the corresponding modeling representative value of the building data after conversion,

is the error value.

It is calculated by a multivariate regression coefficient formula. It should be noted that the linear regression formula is modeled with an administrative district as the smallest unit. If the sample of the administrative district does not meet the sample sufficiency, the model will be modeled with the county and city as the smallest unit. Taking Daan District, Taipei City as an example, the building information is shown in Table 1.

Table 1. Information on buildings in Daan District, Taipei City Building (collateral) information Original value x ₁ : House age 4 years) x ₂ : Number of parks adjacent to 500 meters 6 (pcs) x ₃ : The average transaction unit price of adjacent 300 meters in the past two years 66 (ten thousand yuan) x ₄ : The transaction unit price of the same community in the past three years 70 (ten thousand yuan) x ₅ : Flood depth 1.5 (meters)

Since the original value units of different independent variables are different, in order to be able to compare them together, they are grouped by the value range of the independent variables, and the average house price per square corresponding to each group is calculated as the modeling representative value. Take the age of the house as an example, as shown in Table 2.

Table 2. Conversion of modeling representative value of house age Original value of house age (years) Modeling representative value (average house price per ping, unit: ten thousand yuan) ≤0 85 ≤8 80 ≤14 75 >14 60

The original values in Table 1 are converted as in Table 2, and the information shown in Table 3 is obtained.

Table 3. Conversion of modeling representative values of building data in Daan District, Taipei City Building (collateral) information Original value Modeling representative value x ₁ : House age 4 years) 80 (ten thousand yuan) x ₂ : Number of parks adjacent to 500 meters 6 (pcs) 500,000 yuan) x ₃ : The average transaction unit price of adjacent 300 meters in the past two years 66 (ten thousand yuan) 68 (ten thousand yuan) x ₄ : The transaction unit price of the same community in the past three years 70 (ten thousand yuan) 72 (ten thousand yuan) x ₅ : Flood depth 1.5 (meters) 30 (ten thousand yuan)

（萬元）。淹水後是

（萬元）。則該房價損失率是

。（註：淹水深度0時的建模代表值為50（萬元）。） The unit price per square meter of the building is estimated as follows: Before flooding, it was

(10,000 yuan). after flooding

(10,000 yuan). Then the loss rate of the house price is

. (Note: The modeling representative value when the flooding depth is 0 is 50 (ten thousand yuan).)

，且

，其中，y是應變數，

是迴歸係數，

是該建物資料經過換算後對應的建模代表值，

是誤差值，

是淹水潛勢深度，

是門檻深度。

是由一多變數迴歸係數公式計算而得。以臺北市士林區為例，該建物資料如表4所示。 Referring to FIG. 3, in step 603, when the evaluation device 4 determines that the building data does not have a linear regression, the building data does not have the modeling significance. In step 605 , the estimating device 4 can exclude the building data below a threshold depth from the building data according to an Analysis of Variance (ANOVA) to perform linear regression, and the estimation formula is a threshold regression formula. In step 606, the linear regression formula is used to establish a threshold regression model. The threshold depth is related to the flooding potential depth (eg, with 0.5 m as the tangent point). The threshold regression formula takes the administrative region as the smallest unit for modeling. If the sample of the administrative region does not meet the sample sufficiency, the modeling takes the county and city as the smallest unit. It should be noted that, in step 605, the evaluation device 4 cuts the sample data from shallow to deep according to the flooding potential depth, and judges the flooding according to the variance analysis method under the premise of satisfying the sufficiency of the sample. Whether the water potential depth is significant, after excluding the buildings (guarantees) that do not reach the significant flooding potential depth, the threshold regression model is built using the samples with significant flooding potential depth and other building data as independent variables. mold. The Taiwan-wide inundation potential map is calculated based on the designed rainfall scenarios, specific hydrological and geographical conditions and hydraulic models to simulate the possible inundation of flood control facilities under normal operation. The Taiwan-wide inundation potential map includes the design of rainfall scenarios through rainfall group diagrams and the range of inundation depths marked with different colors according to different inundation depth scales. The definition of the flooding potential depth is to simulate the area that may be flooded through some assumptions (for example, different rainfall), the simulated flooding depth, and the main factor that affects flooding is rainfall. Regression formula at this threshold

,and

, where y is the strain number,

is the regression coefficient,

is the corresponding modeling representative value of the building data after conversion,

is the error value,

is the flooding potential depth,

is the threshold depth.

It is calculated by a multivariate regression coefficient formula. Taking Shilin District, Taipei City as an example, the building information is shown in Table 4.

Table 4. Conversion of modeling representative values of building data in Shilin District, Taipei City Building (collateral) information Original value Modeling representative value x ₁ : House age 4 years) 80 (ten thousand yuan) x ₂ : Floor 5th floor) 35 (ten thousand yuan) x ₃ : The average transaction unit price in the past three years with a radius of 500 meters 57 (ten thousand yuan) 60 (ten thousand yuan) x ₄ : The transaction unit price of the same community in the past three years 70 (ten thousand yuan) 72 (ten thousand yuan) x ₅ : Flood depth 2.0 (meters) 100,000 yuan)

The evaluation device 4 divides the modeling samples into two groups according to the flooding potential depth, the first group is less than or equal to the flooding potential depth, and the second group is greater than the flooding potential depth. The analysis of variance method is used to judge whether there is a difference in the average house price between the two groups of samples, and generate a test value F (referred to as F value) and a significant value P (referred to as P value). If the P value is less than 0.05, the evaluation device 4 judges that the flooding potential depth is significant, which means that there is a difference in the average house price between the two groups of samples, and the flooding potential depth has an influence on the house price.

For example, take the flooding potential depth of 0.5 meters as the tangent point, divide the samples with the flooding potential depth ≤ 0.5 meters into the first group, set one, and divide the rest into the second group , the evaluation device 4 calculates and generates Table 5 according to the variance analysis method as follows.

Table 5. Calculation of ANOVA source degrees of freedom Sum of squared deviations from the mean sum of mean squares F value P value Between groups 1 4235.9 4235.9 14.78 0.0001 s 429 122907.7 286.5 all 430 127143.6

According to the calculation result, the P value is 0.0001, because the P value is less than 0.05, so the evaluation device 4 judges that the flooding potential depth is more than 0.5 meters is significant, which means that in the flooding situation, the flooding depth exceeds 0.5 meters will have an impact on house prices.

（萬元）。淹水後是

（萬元）。則該房價損失率是

。（註：淹水深度0時的建模代表值為35（萬元）。） According to the data in Table 4, the evaluation device 4 judges that the flooding potential depth is 0.5 meters according to the variance analysis method, so the threshold is established after excluding the buildings falling in the area where the flooding potential depth is less than 0.5 meters. regression model. The unit price per square meter of the building is estimated as follows: Before flooding, it was

(10,000 yuan). after flooding

(10,000 yuan). Then the loss rate of the house price is

. (Note: The modeling representative value when the flooding depth is 0 is 350,000 yuan.)

Referring to FIG. 3, in step 606, the estimating device 4 excludes the building data below the threshold depth from the building data according to the variance analysis method and does not have a linear regression, and the estimation formula is the bootstrap method (Bootstrap Method), the threshold depth is related to the flooding potential depth.

Referring to FIG. 3 , in step 607 , the estimating device 4 simulates the distribution of price losses according to the shoe-pulling method and is classified according to regions and building types. The region is divided into north and south-central. The northern part includes Taipei City, New Taipei City, Keelung City, Taoyuan City, Hsinchu County and Yilan County, and the remaining counties and cities are the central and southern regions. The types of buildings are divided into buildings, apartments and houses. In step 608, the estimating device 4 compares the building price in the historically flooded area within a certain period with the average price of the adjacent buildings in the non-flooded area to generate a house price drop, and simulates tens of thousands of times through the shoe-pulling method Then, an allocation of the drop is generated, and the estimating device 4 can estimate the quantile corresponding to the above allocation according to the given probability of the flooding situation.

200公尺為一方格可分為約92萬個方格。參閱圖4，依歷史是否曾有淹水紀錄分為一歷史淹水區及一歷史非淹水區。以北部地區─大廈為例，該估價裝置4計算該歷史淹水區及該非淹水區的房價後，可將樣本整理如表6資訊。 For example, the main island of Taiwan is 200 meters away

A square of 200 meters can be divided into about 920,000 squares. Referring to Figure 4, it is divided into a historical flooded area and a historical non-flooded area according to whether there has been a flooding record in history. Taking the northern area-building as an example, after the evaluation device 4 calculates the house prices in the historical flooded area and the non-flooded area, the sample can be arranged as the information in Table 6.

淹水方格建物B 70 80

淹水方格建物C 50 80

淹水方格建物D 65 80

Table 6. Information on historical flooded areas and non-flooded areas checkered Housing price in flooded area (10,000 yuan) Average house price in adjacent non-flooded areas (10,000 yuan) Price difference (ten thousand yuan) house price loss Flooded Grid Building A 60 80

Flooded Grid Building B 70 80

Flooded Grid Building C 50 80

Flooded Grid D 65 80

The distribution of price losses simulated by the bootstrapping method is shown in Table 7.

Table 7. Simulation Price Loss Allocation Assign quantiles Loss rate corresponding to distribution quantile

Referring to FIG. 5 , after repeating sampling tens of thousands of times in the method of sampling and returning, the evaluation device 4 generates a distribution of house price declines in the northern region-buildings according to the sampling results.

To sum up, the above-mentioned embodiment has the following advantages: Advantage 1, the housing price evaluation device performs model segmentation and distribution simulation based on regions and building types, and can perform different aspects of the danger situation and complete inspection according to the region and building type. It can also be analyzed and compared according to different scenarios to improve the detail of climate change scenario analysis and increase the effect of regional coverage. Achieving the effect is to effectively solve the shortcomings of low detail and coverage encountered in the current practices of various circles.

The second advantage is that in order to evaluate the impact on buildings (real estate collateral) after the flood, the house price assessment equipment is equipped with the probability of scenarios, which can specifically measure the expected loss amount of buildings (real estate collateral) in different scenarios. To achieve efficacy is to achieve the effect of quantifying risk. Therefore, the purpose of this new model can indeed be achieved.

However, the above are only examples of the present invention, which should not limit the scope of the present invention. Any simple equivalent changes and modifications made according to the scope of the patent application for this new model and the contents of the patent specification are still within the scope of the present invention. within the scope of this new patent.

1: Communication interface device 2: Query device 3: Storage device 4: Appraisal device 501～503: Steps of house price assessment 601～608: Steps to determine the applicable model

Other features and effects of the present invention will be clearly presented in the embodiments with reference to the drawings, wherein: FIG. 1 is a block diagram illustrating an embodiment of the novel housing price evaluation device; Fig. 2 is a flow chart of this embodiment; Fig. 3 is the flow chart of a judgment applicable model of this embodiment; FIG. 4 is a distribution map of a historical flooded area and a historical non-flooded area of the embodiment; and FIG. 5 is a graph showing the relationship between a frequency and a house price drop in this embodiment.

1: Communication interface device

2: Query device

3: Storage device

4: Appraisal device

Claims (10)

A house price appraisal device comprising: a storage device having a database, the database records a plurality of building data, and the plurality of building data respectively correspond to a plurality of regions; a query device for receiving a query building signal indicating an area and electrically connected to the storage device, and performing a data lookup table on the database according to the query building signal to generate a query result signal, the query result signal indicating A pair of building data in the corresponding area, the building data includes a flooding information, a number of historical flooded areas related to the flooding information, and the number of historical non-flooded areas, where the historical non-flooded area is the historical the fringe area of the flooded area; and a valuation device connected to the query device for receiving the query result signal from the query device, and generating a house price loss rate according to the building data and an estimation formula, the estimation formula is related to a sample sufficiency, and the sample is sufficient The definition of property is whether the number of pieces in the historical flooded area and the historical non-flooded area related to the corresponding area are respectively greater than or equal to a minimum demand quantity. The house price evaluation apparatus according to claim 1, further comprising a communication interface device electrically connected to the inquiry device, and the communication interface device is used for generating the inquiry building signal according to an input command. The housing price assessment device according to claim 1, wherein the building data further includes a house feature, a geographic information, a public information, a real-price login, and a community information. The housing price evaluation device according to claim 1, wherein the flooding information includes a historical flooding depth and a flooding potential depth corresponding to the area. The housing price evaluation device according to claim 1, wherein the estimation formula is more related to a modeling significance, and the modeling significance is related to whether a linear regression model can be established. The housing price evaluation device according to claim 5, wherein, when the evaluation device determines that the number of pieces in the historical flooded area and the historical non-flooded area corresponding to the area are respectively greater than the minimum demand quantity, and the building data has a linearity regression, then it is judged that the building data has the modeling significance, and the estimation formula is a linear regression formula

, where y is the strain number,

is the regression coefficient,

is the corresponding modeling representative value of the building data after conversion,

is the error value. The housing price evaluation device according to claim 5, wherein when the evaluation device determines that the number of pieces in the historical flooded area and the historical non-flooded area corresponding to the area are respectively greater than or equal to the minimum demand quantity, and the building data is not linear Regression, then it is judged that the building data does not have the modeling significance, but the building data below a threshold depth can be excluded by a variance analysis method to have a linear regression, then the estimation formula is a Threshold regression formula

,and

, where y is the strain number,

is the regression coefficient,

is the corresponding modeling representative value of the building data after conversion,

is the error value,

is the flooding potential depth,

is the threshold depth, which is related to the flooding potential depth. The housing price evaluation device according to claim 5, wherein when the evaluation device determines that the number of pieces in the historical flooded area and the historical non-flooded area corresponding to the area are respectively greater than or equal to the minimum demand quantity, and the building data is not linear Regression, then it is judged that the building data does not have the modeling significance, and the building data below a threshold depth in the building data are excluded by a variance analysis method and there is no linear regression, then the estimation formula is a In the boot method, the threshold depth is related to the flooding potential depth. The housing price evaluation device according to claim 5, wherein when the evaluation device determines that one of the number of pieces in the historical flooded area and the historical non-flooded area corresponding to the area is less than the minimum required number, it determines the building data Without this modeling significance, the estimation formula is a bootstrap method. The house price evaluation device of claim 1, wherein the minimum required quantity is 30 pieces.

Images