JP2001297141A - Device and method for calculating vehicle price - Google Patents

Abstract

PROBLEM TO BE SOLVED: To highly accurately predict the auction contract price of a used car. SOLUTION: A user inputs the basic data of a vehicle, e.g. the type, grade, color, model year, mileage, and remaining automobile inspection period of the vehicle, from an input part 10c. In addition, the user selects and inputs required auction place and area from the input part 10c. A CPU 10a reads out a coefficient corresponding to the input data by using a table previously stored in a memory 10b and calculates the predicted price of the vehicle by using the coefficient and a prescribed operation expression. Since ratios in which area differences are reflected are stored in the memory 10b, the CPU 10a can calculate a price considering trade in a specific area by predicting the price by using the ratio of each area.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and a method for calculating a vehicle price, and more particularly to a technique for predicting a contract price in an auction of a used car in advance.

[0002]

2. Description of the Related Art Conventionally, there has been known a technique for mechanically evaluating a used car based on various vehicle data.

[0003] For example, Japanese Patent Application Laid-Open No. 4-342061 describes a technique for calculating an evaluation point in an auction for selling a used car from the number of months after registration in the first year, a place requiring repair, a mileage, a car name, and the like. I have. That is, various tables for calculating the points and deductions related to the evaluation points from the vehicle data are stored in the storage device, and these tables are appropriately referred to based on the input vehicle data to automatically evaluate the points. It has been calculated.

[0004] The sale price of a vehicle in an auction is:
Since the evaluation point is determined as a guide, the auction bidding price can be estimated to some extent in advance by using this technique.

[0005]

However, successful bid prices in auctions are not uniform nationwide and often vary from region to region. For example, the popularity of 4WD vehicles and RV vehicles is higher in Hokkaido and Tohoku regions, where climate conditions are relatively severe, than in other regions, and auction bidding prices tend to be higher than in other regions. In the above-mentioned conventional technology, such a difference between regions is not considered, and therefore, there is a problem that it is not possible to more accurately predict auction successful bid prices.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems of the prior art, and has as its object to reflect the price difference between regions and to provide a more accurate vehicle price, for example, the auction bidding price of a used car. It is an object of the present invention to provide an apparatus and a method capable of calculating the value.

[0007]

SUMMARY OF THE INVENTION In order to achieve the above object, the present invention provides means for inputting vehicle data, means for inputting regional data for buying and selling vehicles, and statistical data from past trading data for each area. Means for storing a coefficient calculated based on the vehicle data, means for calculating an expected trading price of the vehicle using the coefficient based on the vehicle data and the regional data, and means for outputting the expected trading price. It is characterized by.

The present invention also provides means for inputting vehicle data, means for inputting an auction site of a vehicle, means for inputting regional data where the auction site is located, and statistical data from auction successful bid price data for each region. Means for storing a calculated coefficient, means for calculating an expected auction successful bid price of the vehicle using the coefficient based on the vehicle data, the auction site, and the regional data, and outputting the expected auction successful bid price And means for performing the operation. Here, the means for storing the coefficient includes a means for storing a first coefficient statistically calculated from past sales data for each auction site, and a means for storing a first coefficient statistically calculated from past sales data for each region. In this case, the means for calculating the expected auction successful bid price may include a first and second coefficient based on the vehicle data, the auction site, and the regional data. Anticipate.

[0009] The present invention also provides a method for calculating an expected trading price of a vehicle. The method comprises the steps of: inputting vehicle data and a trade area; and, based on the vehicle data and the trade area, calculating an expected trade price of the vehicle using a coefficient statistically calculated from past trade data for each trade area. A step of calculating and a step of outputting the expected trading price.

The present invention also provides a method for calculating an expected auction bidding price of a vehicle, comprising the steps of: inputting vehicle data, an auction site and an auction area; Calculating a predicted auction successful bid price of the vehicle using a coefficient statistically calculated from past auction successful bid price data for each region to which the venue belongs; and outputting the expected auction successful bid price. Features.

In the present invention, instead of estimating a vehicle price, for example, a successful bid price in an auction, based on only vehicle data as in the related art, a price difference between trading areas is considered. The price difference between trading areas is statistically calculated from past trading data for each area. The division of the region is arbitrary, and the coefficient may be an absolute value for each region, or may be a ratio based on a specific region. By reflecting the past statistical data for each region in the estimated price, highly accurate prediction is possible.

Calculating the expected price of the vehicle based on the regional data includes correcting or correcting the expected price calculated based on the vehicle data using the regional data.

Although the present invention calculates a vehicle price, the present invention can be used for predicting an amount directly or indirectly related to a price, for example, an evaluation point as a guide when calculating a price. .

[0014]

Embodiments of the present invention will be described below with reference to the drawings. In the present embodiment, the vehicle price is predicted in consideration of the difference between the sales prices for each region. If you know that you will buy and sell used cars at a particular auction site,
Although it is possible to reflect the price difference between regions to some extent by using only the past successful bid data at the auction site (the auction site is located in a specific region, Characteristics will be reflected), and the specific data at that particular auction site may affect expected bid prices. A similar problem may occur even if two or three auction venues existing in a specific area are arbitrarily selected and their past successful bid data are averaged and used. The problem becomes more apparent as the number of past auctions at the selected auction site is smaller. In the present embodiment, the vehicle price is estimated in consideration of such circumstances.

FIG. 1 is a block diagram showing the configuration of a vehicle price calculation device according to this embodiment. The vehicle price calculation device 10 includes a CPU 10a, a memory 10b, and an input unit 10c.
And an output unit 10d. Input unit 10c
Is composed of, for example, a keyboard, a mouse, and a touch switch. The output unit 10d can be configured by a display such as a CRT or a liquid crystal. The memory 10b stores the RO
M, RAM, a hard disk, and the like, and arithmetic expressions and various coefficients used when calculating the vehicle price are stored in the ROM or the RAM in advance. The user inputs vehicle data, regional data that he / she wants to buy / sell a vehicle, or specific auction site data that he / she wants to buy / sell a vehicle from the input unit 10c. The vehicle data includes the model and model (grade), color, optional equipment (air conditioner, power steering, transmission, sunroof), year, mileage, vehicle inspection remaining period, etc. The venue is preferably stored in the memory 10b in advance and displayed in a menu format on the output unit 10d so that the user can select and input. The preset trade-desired areas may be, for example, Hokkaido, Tohoku, Kanto, Chubu, Kinki, China, Shikoku and Kyushu. Of course, a finer classification, such as Tokyo, Nagoya, Osaka, etc., can also be used. These data input from the input unit 10c are
a. The CPU 10a calculates an expected vehicle trading price using a predetermined arithmetic expression based on these data and the coefficients stored in the memory 10b, and outputs the calculated expected selling price to the output unit 10d.

The vehicle price calculation device 10 can be specifically constituted by a computer. Also, the memory 10b
May be stored in another computer (server computer) connected to the network. In this case, the vehicle data, the regional data (region selection data), and the auction site data (auction selection data) supplied from the input unit 10c are transmitted to the server computer via a communication control unit and a communication network (not shown). An expected vehicle price is calculated using a predetermined arithmetic expression and a coefficient, and is supplied to the vehicle price calculation device 10. That is, the CPU 10a and the memory 10b do not necessarily need to be provided in the vehicle price calculation device 10, but may be provided in another computer connected to the network.

FIGS. 2 to 10 schematically show tables of respective coefficients stored in the memory 10b.

FIG. 2 is a table showing a basic price (coefficient) of a new vehicle with respect to a vehicle type. Multiple car types A, B, C ...
The basic price of the new vehicle is stored in a table format in advance. The CPU 10a uses the table stored in the memory 10b to input a coefficient of the new vehicle basic price corresponding to the vehicle type (in this case, the coefficient is the new vehicle basic price itself).
Can be obtained. Of course, the new vehicle basic price table can be set in consideration of other factors (for example, color) in addition to the vehicle type.

FIG. (Or grade)
6 is a table showing a ratio (coefficient) with respect to. Model No. for each model A, B, ... A, B, C... Ratios (coefficients) a1, b1, c1,..., A2, b2, c
2 are determined and stored. The ratio is based on a certain type N
o. For example, a model that is most often bought and sold can be used as a criterion. The vehicle type and model No. are input from the input unit 10c. Is input, the CPU 10a can obtain the corresponding ratio using this table stored in the memory 10b. Of course, the coefficient may be an absolute value instead of a ratio to the reference.

FIG. 4 is a table showing the ratio (coefficient) to the color of the vehicle. For each of the models A, B,..., Silver, black, red,.
, d2, e2, f2, ... are determined and stored. This ratio is also for a certain reference color,
For example, it can be determined based on white. Input unit 1
When the vehicle type and the vehicle color are input from 0c, the CPU 10
For a, the ratio of the color can be obtained by using this table. Of course, the coefficient may be an absolute value instead of a ratio.

FIG. 5 is a table showing the ratio (coefficient) to the desired area for buying and selling. Ratios g1 to n in each of a plurality of preset regions for each vehicle type A, B, ...
1, g2 to n2,... Are set. In the present embodiment, the area is divided into eight areas from Hokkaido to Kyushu as described above. This ratio is, for example, for a certain reference area. The two-dimensional table of regions and vehicle types takes into account the fact that regional disparities differ by vehicle type. When the vehicle type and the area are input from the input unit 10c, the CPU 10a can obtain the ratio corresponding to the area using the table. Of course, the ratio is an example of a coefficient, and it is also preferable to use an absolute value for each region as the coefficient.

FIG. 6 is a table showing the ratio (coefficient) to the auction site. Ratios o1, p1, q1,... At a plurality of venues A, B, C,.
, O2, p2, q2,... Are set. This ratio can also be determined based on a specific auction site, for example, an auction site with the largest number of transactions. When a vehicle type and an auction site are input from the input unit 10c, the CPU 10a can obtain a ratio corresponding to the auction site using the table.

FIG. 7 is a table showing the ratio (coefficient) to the market price. Vehicles are not traded on average throughout the year, but demand or supply may be concentrated during a particular month. This table is for considering such market fluctuation factors. Each model A, B, ...
The ratio r1, s1, in January, February, March,.
t1,..., r2, s2, t2,. When the user inputs a vehicle type and a month in which the user wants to buy or sell a vehicle, C
Using this table, the PU 10a can obtain the ratio corresponding to the month and take the market fluctuation into account. The ratio can be calculated based on a specific month, for example, January. Of course, the coefficient may be an absolute value for each auction venue instead of a ratio.

FIG. 8 is a table showing the residual value rate (coefficient) for the model year. As the age gets older, prices generally fall. Residual value indicates the rate of this decrease.
The residual value rates u1, v1, w1,..., U2, v2, w in the year expressions A, B, C,.
2, ... is determined. When the vehicle type and year are input from the input unit 10c, the CPU 10a can obtain a residual value rate corresponding to the year by using this table.
Of course, the coefficient may be an absolute amount instead of a ratio.

FIG. 9 shows the deduction amount (coefficient) with respect to the traveling distance.
FIG. In general, the longer the mileage, the lower the price. .. deduction points x1, y1, z1,... For the traveling distances A, B, C.
, X2, y2, z2,... Are determined. Input unit 10
When the vehicle type and the mileage are input from c, the CPU 10a
By using this table, a deduction amount corresponding to the traveling distance can be obtained.

FIG. 10 is a table showing a point value (coefficient) for the vehicle inspection remaining period. The longer the vehicle inspection remaining period at the time of trading, the higher the vehicle price. Each model A, B,
The additional points α1, β1, γ1,..., Α2, β2, γ2,... In the remaining vehicle periods A, B, C,. When the vehicle type and the vehicle inspection remaining period are input from the input unit 10c, the CPU 10a can obtain an additional point corresponding to the vehicle inspection remaining period by using this table.

As described above, the table of each coefficient for the input item is stored in the memory 10b.
b predicts an expected price at the time of trading, for example, a successful bid price in an auction, using these coefficients. In particular,
The CPU 10a

[Equation 1] Expected price = (base price of new car × model No. disparity coefficient × color disparity coefficient × regional disparity coefficient × auction venue disparity coefficient × market fluctuation coefficient + option evaluation value) × (year residual value rate-mileage deduction) Sum) + Inspection surplus ... (1) Among the items on the right side, the basic price of the new car and the remaining amount of the vehicle inspection are fixed beforehand, but other items,
For example, the regional disparity coefficient is obtained by statistically processing past trading data.

For example, the regional disparity coefficient shown in FIG.
It is calculated by the hill-climbing method from the past auction successful bid price data in the area. In the hill climbing method,

The coefficient of determination = 1− (Σthe square of the difference between the calculated value and the actual price / Σthe square of the difference between the actual price and the average actual price) (2) The coefficient of determination defined by (2) is as close to 1 as possible. Such a solution is defined as an optimal solution. It should be noted that （in the equation (2) is an integration of all the past sales data that satisfy the conditions in the corresponding vehicle type. What satisfies the condition differs depending on what coefficient is to be calculated. For example, when trying to determine the Chubu region among the regional disparity coefficients, it is limited to the trade data of the Chubu region. If the color difference coefficient is
This is past sales data for the corresponding color.

When determining the regional disparity coefficient, the values of all other coefficients, such as the color disparity coefficient, are fixed, and a coefficient such that the determination coefficient is close to 1 based on past trading data of the region of interest ( For example, a ratio is set. If the ratio of a certain vehicle type in the Chubu area is determined with reference to the Kanto region as an example, a specific ratio, for example, 1.1 is set, and all of the relevant vehicle types in the Chubu region are set. Equation (2) is calculated using the past successful bid price data of the above, and another ratio, for example, 1.3 is set, and the equation (2) is similarly calculated. Adopt the ratio of the two. Then, another ratio is appropriately set again, and the same comparison and selection are repeated to search for a ratio that is finally closest to one.

After the coefficients of all the regions (eight regions in this embodiment) are determined, when determining the auction site disparity coefficient next, the regional disparity coefficient is fixed to the coefficient of the region to which the auction site belongs. Applying all past sales data of the corresponding model at the auction site, (2)
A search is made for an auction site disparity coefficient whose coefficient of determination in the formula is closest to one. Thereafter, for example, when determining the color difference coefficient, the coefficient determined so far is one out of the coefficients.
(For example, in the vehicle type to be calculated, if the regional disparity coefficient is determined to be close to 1 in Kanto, the value of the Kanto region is used to fix the regional disparity coefficient to that value.) Calculate similarly. After that, the same operation is performed, and all coefficients are determined from past trading data (the same vehicle type).

According to the hill-climbing method in which a solution selected at random in this way is compared with another solution to find an optimal solution while leaving a more preferable one, not a local solution (two or more Since there is a possibility of convergence to the next best solution), it is also preferable to introduce a genetic algorithm to ensure convergence to the optimal solution.

When determining the coefficient, it goes without saying that another statistical method may be used instead of the hill-climbing method.

As described above, coefficients are determined and stored in the memory 10b, and an expected price is calculated by using these coefficients and equation (1). For example, when the ratio to the reference area is used as the regional disparity coefficient, the coefficient (absolute value) of the area is used in the case of the reference area, and (the coefficient of the reference area) × (the ratio to the reference area) in other areas. it can.

FIG. 11 is a flowchart showing the overall processing of this embodiment. First, the user operates the input unit 10c.
The vehicle data is input from (S101). This vehicle data is vehicle data that will be referred to at the time of trading as described above, specifically, the vehicle type, model No. (Grade), color, presence / absence of optional equipment, year, mileage, and remaining inspection period. After the vehicle data is input, it is determined whether or not the auction site has been input when buying and selling at the auction site next (S102). If a user wants to buy or sell at a particular auction site,
A desired one auction site is selected from a plurality of preset auction sites and input. As one mode, for example, it is preferable that a plurality of auction sites are displayed on the output unit 10d in a pull-down menu format, and the user clicks on one auction site to input. If the user does not expect to buy or sell at a specific auction site, the user does not enter the auction site. In this case, the CPU 10a sets the auction site disparity coefficient to 1 so that the auction site disparity is not considered (S103).

After the input processing of the auction site is completed, it is next determined whether or not the user has input an area (S
104). If the user wants to buy or sell in a specific area, he / she inputs a specific area. This input also displays a plurality of regions set in advance by, for example, a pull-down menu on the output unit 10d, similarly to the selection of the auction venue,
It is preferable to select and input one area by clicking on it. On the other hand, if the user does not expect to buy or sell in a specific area, no area input is performed. For this reason, it is preferable to add, for example, a menu “None” or “No region selection” to the pull-down menu in the region selection. If the user does not select and input a specific region, the CPU 10a sets the regional disparity coefficient to 1 and does not consider the regional disparity (S10).
105).

When the user enters all the data,
The CPU 10a reads the coefficients corresponding to these input items from the table, and calculates the expected price according to the operation formula shown in Expression (1) (S106). Even in the same vehicle,
Since the successful bid price may differ from one auction site to another, it is possible to calculate an accurate expected price when buying and selling at a specific auction site by considering the difference between the auction sites.

Also, by taking into account not only the auction site but also the regional disparity, it is possible to calculate the price in consideration of the regional disparity, and it is possible to make a more accurate prediction. When only the disparity of each auction site is considered, as described above, there is a possibility that the accuracy of the expected price may be reduced due to specific data in the past sales data at the auction site. In this way, by taking into account the regional disparity, the effect of specific data at a specific auction site within the region can be statistically excluded (the number of data in a region is more overwhelming than the number of data in a specific auction) , The contribution rate of the unique data in a specific auction is reduced), and more accurate price calculation is possible.

Also, it is not assumed that an auction will be performed at a specific auction site, and even for a user who simply wants to auction a specific vehicle type in a certain area, for example, Hokkaido, it is possible to calculate prices in consideration of only regional differences. Therefore, the usability of the user is significantly improved.

After calculating the expected price as described above,
The CPU 10a outputs the calculated expected price to the output unit 10d such as a display (S107). The user can know the successful bid price in the auction in advance by looking at the output expected price, and can easily make a trading plan such as whether or not to be placed in the auction or to be placed in a specific area. it can.

FIGS. 12 and 13 show examples of the output screen. FIG. 12 shows a screen when data of a certain vehicle and an auction site are specified (no area is specified) and an expected successful bid price is calculated. FIG. It is a screen in the case of calculating the expected successful bid price in the case where it is specified. FIG.
Is calculated as 1,440,000 yen based on the past data of the corresponding auction site, and in FIG. 13, it is further calculated as 1,480,000 yen based on the ratio of the central region. The actual auction bidding price of this vehicle is 1,483,000 yen, and the applicant has confirmed that the price can be more accurately predicted by taking into account regional differences.

[0041]

As described above, according to the present invention,
The vehicle price can be easily and accurately calculated. Also, for users who expect to buy or sell in a specific area,
It is possible to provide an accurate expected price.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram of an embodiment.

FIG. 2 is an explanatory view (1) of a table stored in a memory.

FIG. 3 is an explanatory diagram (part 2) of a table stored in a memory;

FIG. 4 is an explanatory view (3) of a table stored in a memory;

FIG. 5 is an explanatory view (No. 4) of a table stored in a memory;

FIG. 6 is an explanatory view (No. 5) of a table stored in a memory;

FIG. 7 is an explanatory view (No. 6) of a table stored in a memory;

FIG. 8 is an explanatory diagram (No. 7) of a table stored in a memory.

FIG. 9 is an explanatory view (8) of a table stored in a memory;

FIG. 10 is an explanatory view (No. 9) of a table stored in a memory;

FIG. 11 is an overall processing flowchart of the present embodiment.

FIG. 12 is an explanatory diagram of an output screen according to the embodiment without considering regional disparities.

FIG. 13 is an explanatory diagram of an output screen in consideration of regional disparity in the embodiment.

[Explanation of symbols]

10 Vehicle price calculation device, 10a CPU, 10b memory, 10c input unit, 10d output unit.

Claims (4)

[Claims] A means for inputting vehicle data; a means for inputting regional data for buying and selling vehicles; a means for storing a coefficient statistically calculated from past trading data for each area; A vehicle price calculation device, comprising: means for calculating an expected sales price of the vehicle using the coefficient based on regional data; and means for outputting the expected sales price. 2. A means for inputting vehicle data; a means for inputting an auction venue of a vehicle; a means for inputting regional data in which the auction venue is present; Means for storing a coefficient calculated based on the vehicle data, an auction site, and regional data, means for calculating an expected auction successful bid price of the vehicle using the coefficient, and means for outputting the expected auction successful bid price. A vehicle price calculation device comprising: 3. A method for calculating an expected sales price of a vehicle, comprising: inputting vehicle data and a sales area; and statistically calculating past sales data for each sales area based on the vehicle data and the sales area. A vehicle price calculation method, comprising: calculating an expected sales price of the vehicle using the calculated coefficient; and outputting the expected sales price. 4. A method of calculating an expected auction bidding price of a vehicle, comprising the steps of: inputting vehicle data, an auction site and an auction area; and, based on the vehicle data, the auction site and the auction area, an area to which the auction site belongs. Calculating the expected auction successful bid price of the vehicle using a coefficient statistically calculated from past past auction successful bid price data for each vehicle; and outputting the expected auction successful bid price. Vehicle price calculation method.