JP2005174355A5 - - Google Patents

Description

Vehicle resale price analyzer

The present invention relates to a vehicle resale price analysis apparatus that predicts information related to a sale price, a residual price, or a residual value rate of a vehicle before resale using data relating to a resold vehicle.

  In general, the resale price of a vehicle after the expiration of the usage contract is determined empirically based on the year of the vehicle's first year registration, mileage, etc. They are sold, sent to bidders and auctions, or discarded.

However, the expected resale price based on human experience does not necessarily have a clear basis, and the variation in the expected price by the judge is not small. In addition, since accurate resale prediction cannot be performed, useless transportation and loss in auctions and auctions are caused.
On the other hand, for vehicles based on lease contracts, rental contracts, residual value loan contracts, etc., the gain on sale at the time of resale is set as the residual value in advance. Currently.
On the other hand, in the conventional enterprises, sales are far from the actual situation of ordinary profits, and the view that many companies have more subsidiaries and affiliated companies and the larger sales are better. However, at present, how much dividends can be distributed to shareholders is an important factor, and the management information of the company is made public. The importance of information disclosure also applies to corporate “ratings”. By the way, in the leasing business, it is important to provide a new service, and the revenue is fixed by creating a mechanism for obtaining a reliable revenue at a low cost by controlling resale. In other words, setting a competitive residual value is an extremely important factor. If the residual value can be reasonably calculated and the information can be disclosed, it is possible that the vehicle during the contract period can be financed with asset-backed securities. In this way, it is very important to build a system for setting the residual value.

Accordingly, the present invention provides an article resale price analysis system that can objectively predict the sale price of an article before resale from the sale data of an article such as a vehicle that has already been resold without relying on human experience. For the purpose.
It is another object of the present invention to provide a vehicle resale price analysis system capable of obtaining information on an objective expected sale amount and the like regarding a vehicle to be resold.
Another object of the present invention is to provide a residual value profit / loss analysis system capable of obtaining objective residual value profit / loss information at the time of expiration of a contract for a vehicle in a usage contract period.
It is another object of the present invention to provide an asset evaluation system capable of obtaining objective market price information at an arbitrary time point for a vehicle in a usage contract period.
It is another object of the present invention to provide a residual value setting system capable of obtaining objective residual value prediction information regarding a new contract vehicle.
It is another object of the present invention to provide a storage medium capable of acquiring a correlation formula for obtaining an objective sale amount or the like for a new contract vehicle or a table in which the correlation is associated.
It is another object of the present invention to provide a storage medium that can obtain an objective sale amount for a new contract vehicle.
In addition, the present invention can acquire a correlation formula for obtaining an objective sale amount or the like for a new contract vehicle or a table in which the correlation is associated, and a resold vehicle from which the estimated sale amount or the like is based It is an object of the present invention to provide a storage medium that can output information on the information.
It is another object of the present invention to provide a display device that can output information related to a resold vehicle that is the basis of the estimated sale price.
It is another object of the present invention to provide a residual value setting system capable of setting a residual value related to a new contract vehicle.
It is another object of the present invention to provide a residual value setting system capable of setting a residual value related to a new contract article.
Another object of the present invention is to provide a residual value setting system capable of setting a residual value related to a new model.
It is another object of the present invention to provide a vehicle resale price analysis system that can more accurately obtain an expected sale amount for an average vehicle having no special reason.

The vehicle resale price analysis apparatus according to the present invention according to claim 1 is a vehicle type database for storing vehicle type data relating to a newly produced new vehicle type, and an already resold vehicle database for storing data relating to a resale vehicle transmitted from the bidding site system. A vehicle resale price analysis device for calculating an expected sale amount using data relating to resold vehicles resold within a predetermined period from the vehicle type database and the resold vehicle database, and at least the vehicle type Input means for inputting or selecting data relating to name, contract year, contract expiration year, contract period, new vehicle price, mileage at resale, and assessment at resale, and vehicle type input by the input means Name, use contract year, use contract expiration year, use contract period, new car price, mileage at resale, and assessment at resale A first calculation processing means for calculating an expected sale price and an expected residual value rate using a module calculated or updated by the expected sale price calculation means based on the data to be sold, and the vehicle model name input by the input means Based on the data relating to the above, the actual data of the resold vehicle that matches the vehicle specified by the model name is extracted from the resold vehicle database, and the average mileage, the average sale price, and the average sale rate are calculated. A second calculation processing means for calculating a standard mileage, a standard sale amount, and a standard sale rate using a standard deviation from the average mileage calculated from these actual data, the average sale price, and the average sale rate value; The expected sale amount and the expected residual value rate calculated by the first calculation processing means, and the average mileage, average sale price, average sale rate, standard calculated by the second calculation processing means An output means for outputting a travel distance, a standard sale price, and a standard sale rate together with actual data of a resold vehicle that matches the vehicle specified by the model name extracted from the resold vehicle database, and the expected sale price The calculating means includes means for extracting data relating to resold vehicles resold within a predetermined period from the resold vehicle database, and from the extracted data and data of the vehicle type database, a new car price and a sale price or sale price. Based on the means for calculating the correlation with the residual value rate divided by the new vehicle price as the first equation data, the mileage and the sale amount or the sale amount are calculated by the new vehicle price from the extracted data and the data of the vehicle type database. Means for calculating the correlation with the divided residual value rate as the second equation data, the calculated first equation data and the second equation data Means for calculating the first correction equation data from the average travel distance, and means for calculating the second correction equation data from the calculated first equation data, the second equation data, and the average new vehicle price The third equation data shows the correlation between the popularity index data determined from the first correction equation data and the second correction equation data calculated according to the vehicle type data or the vehicle type and the residual value rate or the sale price. And means for calculating as follows .

As described above, according to the present invention, an article resale price that can objectively predict the sale price of an article before resale from the sale data of an article such as a vehicle that has already been resold without relying on human experience. An analysis system can be provided.
In addition, the present invention can provide a vehicle resale price analysis system that can obtain information on an objective expected sale amount and the like regarding a vehicle to be resold.
In addition, the present invention can provide a residual value profit / loss analysis system that can obtain objective residual value profit / loss information at the time of expiration of a contract for a vehicle in a usage contract period.
In addition, the present invention can provide an asset evaluation system that can obtain objective market price information at an arbitrary point in time for a vehicle in a usage contract period.
Further, the present invention can provide a residual value setting system capable of obtaining objective residual value prediction information regarding a new contract vehicle.
In addition, the present invention can provide a storage medium that can acquire a correlation formula for obtaining an objective sale amount or the like for a new contract vehicle or a table that associates the correlation.
In addition, the present invention can provide a storage medium capable of obtaining an objective sale amount or the like for a new contract vehicle.
In addition, the present invention can acquire a correlation formula for obtaining an objective sale amount or the like for a new contract vehicle or a table in which the correlation is associated, and a resold vehicle from which the estimated sale amount or the like is based The storage medium which can output the information regarding can be provided.
Moreover, this invention can provide the display apparatus which can output the information regarding the resale vehicle used as the origin of the estimated sale amount.
The present invention can also provide a residual value setting system that can set a residual value related to a new contract vehicle.
In addition, the present invention can provide a residual value setting system that can set a residual value related to a new contract article.
In addition, the present invention can provide a residual value setting system capable of setting a residual value relating to a new model.
In addition, the present invention can provide a vehicle resale price analysis system that can more accurately obtain an expected sale amount for an average vehicle having no special reason.

The present invention uses the data regarding the actual usage period and the new vehicle price, which is data that has already been empirically recognized from among sales data related to vehicles that have already been resold, that has a large impact on the sales price. Objectively expected sale amount for vehicles scheduled for resale by obtaining a correlation formula or a table that correlates the correlation to obtain information on expected sale value, expected residual value, or expected residual value rate Can be obtained.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
First, in the present invention, the use contract is a contract that permits use or ownership for a predetermined period such as a lease contract, a rental contract, or a residual value loan contract, and the expiration of the use contract means that the lease or rental period ends. It is intended to terminate the contract due to mid-term cancellation.
In the following description of the embodiments of the present invention, a vehicle resale price analysis system based on vehicle data resold at auction or auction after the rental contract expires will be described as an example.

FIG. 1 is a block diagram showing an overall configuration including a vehicle resale price analysis system according to the embodiment.
The resale support system 10 includes a support side vehicle resale system 11 and a bid support system 12. The support-side vehicle resale system 11 includes a vehicle type database 13, an already resold vehicle database 14, and an expected sale amount calculation system 15.
Here, the vehicle model database 13 includes manufacturer name, vehicle model name, vehicle application, vehicle shape, vehicle model grade (vehicle model name, grade name), certified model, common model (model designation number, category classification number), mission, drive system, Displacement, number of doors, capacity, load capacity, engine type (motor model), number of engine cylinders, engine mechanism, tire size, turbocharger, roof shape, exhaust emission regulations, body size, body color, vehicle tax classification, weight It has data such as tax, insurance class, popularity index, vehicle usage, vehicle shape, start of sale, end of sale. Here, the popularity index is an index that is ranked by the residual value rate in the classification classified by vehicle type. For example, if the residual value rate of Corolla and Civic are the same, and the residual value rate of Sunny is 2 rank lower than Corolla, the popularity index of Corolla is set to 25, the popularity index of Civic is set to 25, and the popularity index of Sunny is set to 22. Is done. The vehicle use is a classification of vehicles according to the use, and is classified into passenger cars, vans, buses, trucks, and the like. The vehicle shape is classified according to the number of doors and the outer shape. For example, a sedan (SD) if 4 doors + trunk, a hatchback (HB) if 2 doors + tailgate (no 4 door specification), 2 doors. + Trunk coupe or sports (CP), 2-4 door + tailgate or 4 door base full bonnet, bonnet wagon (BW), 3-4 door + tailgate or semi cab over cab wagon (CW).
The resold vehicle database 14 also relates to resold vehicles such as the use contract year, the expiration date of the use contract, the use contract period, the new car price, the sale price after the use contract expires, the mileage at the time of resale, and the assessment evaluation at the time of resale. I have data.
Further, the expected sale price calculation system 15 acquires a multiple regression formula, a correlation formula, or a table that associates the correlation from the data of the vehicle type database 13 and the already resold vehicle database 14, and calculates the expected sale price before resale and the expected remaining price. This is a system for calculating information on value or expected residual value rate.
The bidding system 12 is a system for selecting and submitting domestic bidding, overseas bidding, bidding on the web using the Internet, auction, retail at a store, and the like. For example, the bidding system 12 transmits the exhibition data regarding the exhibition vehicle to the bidding site system 16 of the bidding site.

The use side system 20 used by a leasing company or the like has a use side vehicle resale price analysis system 21. The user-side vehicle resale price analysis system 21 includes a residual value setting system 22 that sets a residual value related to a new contract vehicle, and a residual value simulation system 23 that performs residual value simulation. Here, as the residual value simulation system 23, for example, a residual value profit / loss analysis system 24 that predicts residual value profit / loss of a vehicle before resale, or an asset evaluation system 25 that predicts a market price at an arbitrary time point regarding a vehicle during a use contract period. There is.
In addition to the use side vehicle resale price analysis system 21, the use side system 20 includes an estimation system 26 and a backbone system 27 that are used at the time of lease contract. Here, the basic system 27 includes a lease contract database 28 for storing lease contract data. The user side system 20 has an exhibition support system 29. The exhibition support system 29 is a system that transmits contract expiration data to the bidding system 12 regarding a vehicle whose contract has expired, or preferably a vehicle whose contract is scheduled to expire after a predetermined period.

Next, the flow of exhibition data to a bid party will be described.
Contract expiration data relating to a vehicle whose contract has expired, or preferably a vehicle whose contract is scheduled to expire after a lapse of a predetermined period, is sent from the backbone system 27 to the exhibition support system 29. In the exhibition support system 29, the contract expiration data is transmitted to the use side vehicle resale price analysis system 21. The use side vehicle resale price analysis system 21 transmits the expected sale price data based on the latest data to the exhibition support system 29. The exhibition support system 29 that has received the expected sale price data transmits the exhibition data including the expected sale price data (desired sale price) in the contract expiration data to the bidding system 12. The bidding system 12 selects an optimal sales destination from domestic bidding, overseas bidding, bidding on the web using the Internet, auction, retail at a store, etc. based on the received exhibition data. When domestic bidding is selected, the exhibition data is transmitted to the bidding site system.

Next, the flow of resold vehicle data that has been bid will be described.
The tendered resale vehicle data is sent from the tender hall system 16 to the already resold vehicle database 14. The resale vehicle data sent to the already resold vehicle database 14 is regularly used in the expected sale amount calculation system 15 and is periodically transmitted as update data to the use side vehicle resale price analysis system 21.

Next, data update in the vehicle resale price analysis systems 10 and 21 will be described.
First, the vehicle type database 13 additionally updates vehicle type data related to a new model every time a new model is newly produced or produced. Here, the new model includes the case where the model certification number is changed. The data stored in the vehicle type database 13 is regularly used in the expected sale amount calculation system 15 and is transmitted as update data to the use side vehicle resale price analysis system 21 periodically or when data is updated. Further, the expected sale amount calculation system 15 is also periodically updated with new data, and the updated module such as the correlation formula is sent as update data to the use side vehicle resale price analysis system.
The data in the lease contract database 28 is used for various analyzes in the residual value simulation system 23.

Next, an embodiment of a method for obtaining a residual value calculation equation used for analyzing a vehicle resale price will be described with reference to FIGS.
FIG. 2 is a processing flow showing a method for acquiring a residual value calculation equation.
First, as a first step, sales performance data relating to already resold vehicles is prepared, and predetermined data relating to already resold vehicles is extracted (S1).
The data extracted here is data relating to resold vehicles such as vehicle type data, use contract year, use contract expiration year, use contract period, new vehicle price, sale price after use contract expiration, mileage at the time of resale, and the like. Here, the vehicle type data is data for specifying a vehicle such as manufacturer name, vehicle type name, certified model, vehicle shape, displacement, fuel, shift, drive system, number of doors, grade, and the like. The usage contract year, the expiration date of the usage contract, and the usage contract period are data determined by the usage contract, and not all these data are required. For example, together with other data such as the first year registration, Any data that can determine the contract period, contract time, or contract expiration time may be used. In addition, data such as vehicle model name, certified model, displacement, vehicle shape, fuel, shift, drive system, number of doors, equipment and other grades and grade options are attached to the vehicle to determine weight tax. Since it can be estimated from the model designation number and the category classification number, the model designation number and the category classification number can be used as data instead of these individual data. Moreover, it is preferable to include data such as the vehicle body number, the first year registration, the registration number, and the vehicle inspection deadline date described in the vehicle verification. The new car price is a standard selling price at the time of a new car. When the standard selling price varies depending on the region, the standard selling price in the reference region is used, but regional disparities may be considered. Further, not only the standard selling price but also the actual selling price can be used. In addition, when various equipment such as an air conditioner and a navigation system is provided at the time of delivery of the new vehicle, the vehicle value varies depending on these equipments. Therefore, it is preferable to treat the selling price including these equipment as the new car price.
The predetermined period for extracting data is determined in consideration of economic conditions, market trends, product cycles, data parameters, and the like. That is, it is preferable to set a shorter period for economic and market trends if the fluctuation is large. As for the product cycle, if the cycle period is long, the period may be set longer. It is important that there is a sufficient number of data parameters for statistical processing. For example, sale data whose sale date (bidding date) is within the past two years is extracted.
In addition, the actual use data relating to the article is data relating to the use state relating to the target article. This actual use data includes user classifications such as lease and rental even for individuals, corporations and corporations. In addition, the actual use data may be data relating to the use period, use conditions, use application, or equipment and accessory equipment added during the use period. For example, in the case of a personal computer or equipment, the presence or absence of software for operating these devices has value as actual use data.

Next, as a second step, data normalization is performed (S2). First, in order to normalize the data, the data bias is corrected. Randomly extract data to eliminate data skew. If there is a bias, theoretically correct the data selection or add data in consideration of the cause of the bias.
In addition, for the normalization of data, a bidet car and an unlisted car are considered. In other words, after the expiration of the period of use, betting or unlisted vehicles that could not be resold are considered as risks and excluded (deleted) from the data target. In addition, it is good to collect the tag data of the tag car and the unlisted car for each vehicle type, and calculate the tag price rate for each product characteristic and each market.

  Next, as a third step, a residual value table is set in consideration of product characteristics and a market (S3). That is, already resold goods are distinguished according to the characteristics of the product or the market. In the present embodiment, a vehicle is classified according to vehicle use such as passenger, commercial, cargo, or bus, or a vehicle shape such as a sedan type, hatchback type, or one-box type, and one division is selected. Then, for example, data in which the vehicle usage is classified as riding is extracted.

  Next, as a fourth step, an analysis target period to be analyzed is set (S4). For example, it classifies by lease period and divides into several categories. Then select several periods with a large data parameter as representative periods. If there are many variations in the usage period or if it is continuous, an appropriate period is selected in consideration of each category or product cycle. For example, select 3, 4 or 5 year lease expired cars from the past sale data, and for those 3 year lease expired cars, data with an elapsed time of sale of 35 to 37 months is extracted as a representative period. .

Next, as a fifth step, among sales data, items that can be known at the time of sale or items that can be set are subjected to correlation analysis at various angles, and those having positive / negative correlations are picked up to create a scatter diagram (S5). ).
For vehicles, the elapsed period or usage period from the first year registration, the popularity index determined according to the vehicle type or vehicle type, the new vehicle price, the amount sold at the time of resale, the mileage at the time of resale, and actual use other than the mileage Data can be picked up as items and the correlation between them can be taken into account. Regarding vehicles, the relationship between the new vehicle price and the residual value ratio obtained by dividing the sale amount or sale amount by the new vehicle price, the relationship between the mileage and the residual value rate obtained by dividing the sale amount or sale amount by the new vehicle price, and the vehicle type It was possible to obtain from the actual analysis that the correlation between the popularity index data determined according to the data or the vehicle type and the residual value rate or the sale price has a high correlation.
Here, with respect to the articles, the sales period or use period, the selling price, the sale amount at the time of resale, and the actual use data at the time of resale can be picked up as items, and the correlation between them can be considered. Among these items, the correlation between the selling price and the sale price or the residual value ratio obtained by dividing the selling price by the selling price, the actual use data and the residual value ratio obtained by dividing the selling price or the sale price by the selling price, and It is preferable to consider the correlation between the vehicle price data or the popularity index data determined according to the vehicle type and the residual value rate or the sale price.
Scatter charts created in this way are shown in FIGS.
In this embodiment, using the data extracted in the fourth step (S4), the X-axis is the new car price, the Y-axis is the sale price, the scatter diagram shown in FIG. The scatter diagram shown in FIG. 4 is created as the value rate.

Next, as a sixth step, an approximate curve (normal equation) is created from each scatter diagram and the tendency is converted into a function (S6).
In the present embodiment, approximate curves (normal equations fa (x), fb (x)) are created from FIGS. 3 and 4, respectively, and the trend is functionalized to obtain equation data.
Here, fa (x) is equation data indicating the correlation between the new car price and the sale price, and fb (x) is the correlation between the mileage and the residual value rate obtained by dividing the sale price by the new car price. It is the equation data which shows a relationship.

Next, as a seventh step, each normal equation is used to mutually correct the base to the average value to obtain correction equation data (S7). With this correction equation data, mutual effects are removed from each acquired normal equation, and a more accurate normal equation with respect to XY correlation is acquired.
Specifically, first, the average new vehicle price and average travel distance of the entire data C3 are calculated. Then, using the average new car price and the average mileage, the average distance conversion sale amount and the average new car price conversion sale residual value ratio are obtained by the following formula.
Average distance converted sale amount = fa (new car price)-fb (actual mileage) * [new car price] + fb (average mileage) * [new car price] Average new car price converted sale residual value ratio = fb (actual mileage)- fa (new car price) / [new car price] + fa (average new car price) / [new car price]
Then, the scatter diagram of Fig. 5 is created with the X-axis as the new car price, the Y-axis as the average distance converted sale price, and the scatter diagram of Fig. 6 as the X-axis as the travel distance and the Y axis as the average new car price converted sale residual value ratio. Then, approximate curves (normal equations fa ′ (x), fb ′ (x)) are created by regression analysis for the correlation of the average values from FIGS. 5 and 6, and the tendency is converted into a function.
Here, the normal equation fa ′ (x) is correction equation data in consideration of the average travel distance, and fb ′ (x) is correction equation data in consideration of the average new vehicle price.

Next, as an eighth step, using the normal equation obtained in the seventh step, a standard expected sale price is obtained by calculation from the new vehicle price and travel distance of the sale data (S8).
The standard expected sale price can be obtained by the following formula.
Standard expected sale price = fa '(new car price) + fb' (actual mileage) * [new car price] -k
However, k is a distance constant determined by the following.
k (distance constant) = fb (average mileage) * [new car price]
Although the standard expected sale price can be used as the expected sale price, more accurate forecast sale price can be acquired by further classifying the product by characteristics / market.

Next, as a ninth step, a function for obtaining the expected sale amount is obtained (S9).
First, the product is divided into groups according to the characteristics / markets of the products to form groups in the residual value table. Then, the difference between the actual sale amount and the theoretical standard expected sale amount is acquired, and the residual is reflected by group. In other words, the value or popularity brought about by the characteristics of each group that cannot be compensated only by the analysis results is taken into consideration.
Specifically, the theoretical standard expected sale price / new car price is subtracted from the actual sale price / new car price, and the difference between the actual sale residual value ratio and the theoretical sale residual value ratio is obtained and averaged for each vehicle type. Rank in descending order in the positive direction.
That is, first, the individual (sales amount−standard expected sale amount) / new vehicle price = RZ of the data extracted in the fourth step is obtained. Here, let RZ be the deviation, and let Z be the sale amount−the standard expected sale amount.
Then, the standard deviation (HRZ) of RZ by vehicle type by vehicle shape of the data extracted in the fourth step is obtained. And (RZ + HRZ) is calculated | required from (RZ-HRZ) for every vehicle type vehicle shape, and the average deviation (ARZ) of RZ (deviation) is calculated | required. Thereafter, the groups according to vehicle shape are indexed in order of increasing average deviation (ARZ) in the positive direction, and popularity index data is acquired.
Then, a scatter diagram shown in FIG. 7 is created using the popularity index data with the rank as the X axis and ARZ as the Y axis, and a normal equation (fc (x)) is obtained by regression analysis for the correlation of the average values.

Next, the normal equation is adjusted as a tenth step (S10). The adjustment of the normal equation is performed by determining the difference in the degree of wear caused by a factor that is not known at the time of sale or a factor that cannot be set, by using the sales amount / new car price variance of the same group.
Specifically, it is assumed that the standard deviation = the difference in the degree of consumption≈assessment, and when the degree at the time of return is assumed in advance, the deviation value and the assessment score are linked and reflected in the residual. If not, consider risk hedging and rank down according to certain criteria (Figure 8).
That is, assuming that the standard deviation = the difference in the degree of wear≈assessment, for example, if the assessment score 2 is defined as the deviation value 45, the value given from the residual value rate (fc (x)) of the deviation value 45 from the expected assessment score 2 Can acquire the deviation value 50). Further, if the deviation value 47 is defined as a standard for risk hedging as a whole, the profit for the three deviation value sections is theoretically secured. This risk hedging process is adjusted by lowering the rank so that the theoretical average selling price ≦ the actual average selling price.

    Next, as an eleventh step, the necessity of selecting another predetermined period is determined (S11). That is, when it is necessary to earn a function for a representative period different from the already selected category, the process returns to step 4 to set the representative period again. If it is not necessary to set another representative period in the category of the remaining value table that is currently selected, the process proceeds to step 12.

  Next, as a twelfth step, the necessity of setting another category is determined (S12). If it is necessary to change the category of the remaining value table that has already been selected and obtain a function for another category, the procedure returns to step 3 to set another category again. Here, if it is not necessary to set another category, the acquisition of equation data is terminated.

When it is necessary to add a new model, compared to the existing model, the popularity index of the nearest model is calculated based on the model or popularity index that seems to be the same class from the model usage and the new car price. Apply. As a specific method, when a new model is added, the popularity index of the group by vehicle type by vehicle type assumed to be equivalent is applied with reference to HRZ or ARZ by vehicle type by vehicle type. The new model includes a model change when the model certification number is changed.
In addition, the expected standard sale amount for an arbitrary period is calculated by obtaining the expected standard sale amount from the representative period statistically analyzed in the nearest period before and after this arbitrary period, and determining the equation to be proportional during that period. . When the representative period is only on either side of the arbitrary period, two representative periods are selected from the side closer to the arbitrary period on the side where the representative period exists, and the three periods are calculated to be proportional.
Then, regarding the sampled period, the estimated standard sale price can be output if conditions are picked up by the residual value table and subjected to regression analysis.
Specifically, if the data such as new car price, expected mileage, car type (rank), etc., for the sampled period is given to the normal equation that is picked up in the residual value table group and regression analyzed, the expected sale price can be obtained. it can.
The expected standard sale amount can be obtained by the following formula.
Expected standard sale price = fa ′ (new car price) + fb ′ (actual mileage) * [new car price] −k + ARZ of vehicle type by vehicle type group * [new car price] + HRZ / 10 * of vehicle type by vehicle type group ((Designated deviation value) -50) * [New car price]

In addition, it is preferable to consider the reason for showing a certain tendency.
The vehicle has a special feature that the evaluation depends on the year. Therefore, in order to predict the resale rejection amount for the vehicle, it is important to reduce the amount in consideration of the peculiar age. For example, if the first year registration sells a 3-year lease expired vehicle in December in the following month, the vehicle will be 4 years old. In this way, it is considered that there is a slight decrease in the sale price for articles that are often evaluated by year.
In addition, since the expected standard sale price is pure sale price, the increase / decrease amount should be determined by adding the selling expenses, indirect expenses such as profit on sale, or strategic profits such as strategic products as the residual value. Is preferred.
Moreover, it is preferable to increase or decrease the price by converting the deviation value by using the variance of the data by group and the standard deviation as a risk hedge of the expected standard sale price.

Next, another embodiment of a method for acquiring a residual value calculation equation used for analyzing a vehicle resale price will be described with reference to FIG.
FIG. 9 is a processing flow showing a method for acquiring a residual value calculation equation. The same steps as those in the embodiment shown in FIG.
In the fifth step, a scatter diagram may be created in the same manner as in the above embodiment, but an item may be selected by performing a correlation analysis on data that seems to affect the residual value rate.

In the sixth step, multiple regression analysis is performed on the items with hits, the correlation is confirmed, and appropriate items are selected (S16).
As a result of extracting sales data (sale date) within the past two years and performing multiple regression analysis, the number of months elapsed, monthly mileage, new car price, displacement, light automatic classification (660CC or less or not) ), Automobile tax classification (luxury car = 3000CC or not), vehicle usage classification (passenger or not), new and old classification (elapsed months or less or not), fuel (gasoline or not) ), I was able to find a correlation as to whether or not it is equipped with ABS. In addition, the specific vehicle type classification (in this case, a classification that classifies whether the vehicle type / vehicle shape shows a special tendency such as a Corolla or a sprinter, which is particularly popular, for example) is also shown.
Here, the number of elapsed months, monthly mileage, new car price, and displacement are independent variables, and “1” for light automatic classification, automobile tax classification, vehicle shape classification, old and new classification, fuel classification, and specific vehicle classification. Or a dummy variable of “0”.
Regression analysis of each selected item and the residual value rate is performed, and data is indexed (indexed by logarithm, power, index, etc.) so as to be applicable to linear regression in some cases. For example, the number of months elapsed is indexed logarithmically.
Then, (raw data−average value) / standard deviation = standardized data, and the data is statistically standardized data.
Then, a correlation between the selected item and the residual value rate is obtained by multiple regression analysis, and a partial regression coefficient and an intercept (constant term) numerical value are obtained for each item.

As a seventh step, whether or not the result is reliable is evaluated by a determination coefficient, a multiple determination coefficient, a t-test or the like (S17).
In the eighth step, the partial regression coefficient and the intercept (constant term) are applied to the multiple regression equation, the theoretical residual value rate is obtained from the selected item, and the difference from the actual residual value rate is obtained as the residual (S18). ).
As a ninth step, a residual value rate is determined (S19).
Obtain the average and standard deviation of residuals for each category and use them as the theoretical residual value rate. The standard deviation is defined as the variable factor of the sales amount due to factors that cannot be predicted in advance, or the variable factor of the sales amount due to factors that cannot be grasped at present. Combined with risk hedging, the residual value rate is determined by adding or subtracting to the theoretical residual value rate as an estimated difference in wear level = assessment.
For example, the difference in the degree of wear is defined in five stages (1 = deviation value 40, 2 = deviation value 45, 3 = deviation value 50, 4 = deviation value 55, 5 = deviation value 60), and the predicted assessment score is calculated. The residual value rate is obtained by adding or subtracting the residual value rate to the theoretical residual value rate from the predicted assessment score with a deviation value of 50.
Note that the case of adding a new model is the same as in the above embodiment, and the description thereof is omitted.

Using the multiple regression formula obtained by the above steps, popularity index and standard deviation by category, vehicle type database, identification of vehicle type from vehicle type database, number of months elapsed, contract monthly mileage, new vehicle price, estimated assessment score Can be used to calculate the sale price after the contract expires.
The popularity index and standard deviation by category, light vehicle classification, automobile tax classification (luxury car), and vehicle shape classification can be obtained from the vehicle type database, and the old and new classification should be obtained from the number of months elapsed in the contract by calculation processing. Can do.
FIG. 10 shows a multiple determination index for each item considered as a result of a multiple regression analysis by extracting sales data (the number of data is 35,000) within the past two years. As preconditions, light automatic classification, automobile tax classification, vehicle usage classification, old and new classification, fuel classification, and specific vehicle classification are considered as dummy variables. The selected items are elapsed months, displacement, new car price, and monthly mileage.
Example 1 considers all elapsed months, displacement, new vehicle price, and monthly mileage, Example 2 considers elapsed month, new vehicle price, and monthly mileage, Example 3 Example 4 considering elapsed months, displacement and monthly travel distance, Example 4 considering elapsed months, displacement and new car price, Example 5 considering displacement, new car price and monthly travel Example 6 considers distance, Example 6 considers elapsed months and new vehicle price, Example 7 considers new vehicle price and monthly mileage, Example 8 shows elapsed months and displacement In consideration, Example 9 considers the displacement and the monthly travel distance.
As shown in FIG. 10, Example 1 considering four items has the highest matching rate, but Example 2 shows a matching rate close to Example 1 regardless of three items.
Moreover, Example 3, Example 4, and Example 6 also have the next highest result of the multiple determination index. In particular, Example 6 shows a high coincidence rate despite having two items.

Next, an analysis system for sale price, residual value profit / loss, asset prediction, and determination of the residual value of a new contract vehicle will be described with reference to FIGS. FIGS. 11 to 18 are screen image diagrams of the system.
FIG. 11 is a screen image diagram for obtaining an expected sale amount and an expected residual value rate for, for example, a specific vehicle during a contract period or a specific vehicle at the time of a new contract.
In the figure, the model name “Corolla”, specification “Diesel DX 4FAT 2WD”, lease period “60” months, scheduled registration date “00/05/15”, expected mileage “100” 1,000 km, rating “3”, And the new car amount “1,272” thousand yen, the expected sale amount “191” thousand yen, the standard sale amount “197” thousand yen, the average sale amount “186” thousand yen, the expected residual value ratio “15. 0% ", standard sale rate" 15.6% ", average sale rate" 14.7% ", standard sale amount and standard sale rate," 12 units "already sold, this number of already sold vehicles The standard distance of “12” is “102” thousand km, the average resale vehicle number and the average sale rate are “18”, and the average distance of this resale vehicle number “18” is “101” thousand. km and past vehicles sold (successful bid, sale rate, manufacturer, vehicle type, Both applications, the exhaust amount, grade, number of months, year, new car price, mileage, transmission, fuel, drive, and outputs display to hold month and year).
The vehicle type name “Corolla” and the specification “Diesel DX 4FAT 2WD” can be selected and input by a pull-down method. Also, in this figure, “Corolla” and the specification “Diesel DX 4FAT 2WD” are entered to display the manufacturer name “Toyota”, certified model “KA-CE106V”, vehicle shape “BV”, and displacement “2000”. However, instead of inputting "Corolla" and specification "Diesel DX 4FAT 2WD", the manufacturer name "Toyota", certified model "KA-CE106V", vehicle shape "BV", and displacement "2000" are input. It may be. Moreover, it may replace with the input of "Corolla" and specification "diesel DX 4FAT 2WD", and may input a model designation number and a classification classification number.
Further, the estimated travel distance “100” thousand km may be linked to the lease period “60” months without being input. The new car amount “1,272” thousand yen is also an item that can be determined depending on the car type and specifications, and can be output and displayed from a database associated with the car type name in advance. The rating “3” is an assessment evaluation, and is classified according to, for example, a user such as a leaser, a place of use, or a purpose of use (commercial or private use).
The “expected sale price” and “expected residual value ratio” are calculated and output using equation data and multiple regression equations obtained in advance in the above embodiment. On the other hand, the “average sale price”, “average sale rate”, and “average distance” are calculated and output from actual data of resold vehicles that match the vehicle specified by the “model name”. In the case of the same figure, it shows that there are 18 corresponding vehicles. On the other hand, “standard sale amount”, “standard sale rate”, and “standard distance” are obtained by correcting the data bias as (actual data−average value) / standard deviation. In the case of the figure, it shows that there are 12 corresponding vehicles after correction.
The output is sufficient to indicate the expected sale amount or the expected residual value rate, but by showing both the expected sale amount and the expected residual value rate, there is an effect that it is easy to grasp the residual value and the like. In addition, by showing the standard sale price, average sale price, standard sale rate, average sale rate, etc., it is easy to grasp the accuracy of the expected sale amount and the expected residual value rate, and also grasp the presence or absence of specificity by vehicle type be able to.
In addition, past sales vehicles (successful bid amount, sale rate, manufacturer, vehicle type, vehicle shape, displacement, grade, number of months, year, new vehicle price, mileage, mission, fuel, drive, date held) By doing so, it is possible to confirm a factor that becomes higher or lower than the average value.

FIG. 12 is a screen image diagram for predicting residual value profit / loss based on the expected residual value, for example, for a specific vehicle during the contract period.
In this embodiment, the market at the time of expiration of the contract is predicted based on the current market, and a relative value is set by relative evaluation. Then, an expected standard sale amount at that time is obtained from unsold contract data, and a relative value at the time of contract expiration is multiplied to obtain an expected standard sale amount at the time of contract expiration. Then, the residual value profit / loss is obtained by the contract residual value and the expected standard sale amount.
Specifically, the expected standard sale amount at that time is calculated from the new vehicle price, estimated mileage, model name, and (scheduled assessment score at the time of return) of unsold contract data, and the contract expires by multiplying the relative value at the time of contract expiration The expected standard sale amount at the time of Then, based on the contract residual value minus the expected standard sale amount, the management unit and the product characteristic / market by vehicle type and the vehicle type are summed up to determine the residual value profit / loss for each management unit based on the accounting unit.

In the figure, “contract residual value”, “expected residual value”, and “remaining profit / loss” for each vehicle type and year are displayed for the model name “Corolla” currently under contract.
Here, “contract residual value” is the residual value set at the time of contract, and “expected residual value” is the residual value calculated and output using the equation data or multiple regression equation obtained in advance in the above embodiment. is there. “Residual value gain / loss” is the difference between “contract residual value” and “expected residual value”. If it is close to zero, it means that there is a prospect of sale according to the residual value set at the time of contract, and no profit or loss has occurred. To do.
For example, “Corolla BV” is expected to generate a profit of “968” for a vehicle whose contract expires in 2000, but a loss of “9039” for a vehicle whose contract expires in 2003 It shows that it becomes a prediction.
Although the vehicle type display is shown in the figure, it is possible to display all contracted vehicles or display by a specific manufacturer name. In addition, by registering the dealer category and the salesperson category in the database, it is possible to output the residual value profit / loss for each dealer and each salesperson. Although omitted in the figure, the profit / loss per vehicle can be known by displaying the number of target vehicles.
FIG. 13 is a screen image diagram for predicting residual value profit / loss for each vehicle type for the vehicle in the contract period. As shown in the figure, residual value gains and losses are displayed for each vehicle type and vehicle shape. In this way, by displaying the residual value profit / loss by vehicle type and vehicle shape, it is possible to compare the profit / loss status of each.

FIG. 14 and FIG. 15 are screen image diagrams for predicting residual value profit / loss based on the expected residual value for a specific vehicle during the contract period, for example, and are substantially the same as FIG.
14 and 15 are characterized in that they are systems capable of dealing with economic fluctuations. “100%” is displayed in the upper column of the year display column from “2000” to “2006”, but when all “100%” is displayed, it indicates that the business fluctuation is not taken into consideration.
For this economic fluctuation, it is possible to display the fluctuation rate based on other economic situation analysis data, etc., and to apply it to each expected residual value.
In addition, by giving future fluctuation factors to the residual value such as future market fluctuations and fluctuations in sales volume, etc., how the expected standard sale amount varies for each residual value group (management unit, model) You can also simulate what will happen. In such simulation, it is preferable to perform simulation including the future by giving a planned sales amount (budget).
The optimization direction can be determined by the above simulation.
In other words, it is possible to decide whether to set the residual value profit / loss to the specified amount, to secure the specified value for the residual value profit / loss, to collect the fixed rate to the contract new car price, to collect the fixed rate for each contract, or to collect the fixed amount for each contract. .
In addition, when a residual value group or model (vehicle type) is specified within a range that gives the expected sales volume for each new vehicle price, the new vehicle price and the expected sales volume are given for each specified residual value group or model (vehicle type). If the new car price and the expected number of units sold can be given for each residual value group or model (model), the accuracy will increase. If sales prospects are not given, sales results can be substituted.
By calculating after setting the above conditions, the expected standard sale amount + adjustment amount is set as the residual value for each management unit, residual value table, residual value group or model (vehicle type). It is preferable to take into account risks, profits, overhead costs, etc. that are recalculated for each management unit period.

16 to 18 are screen image diagrams for obtaining the overall tendency of the expected sale amount and the expected residual value rate. FIG. 16 is a screen image showing trends in new car prices and successful bid prices by lease period, FIG. 17 is a screen image showing trends in mileage and residual value rates by lease institution, and FIG. 18 is lease periods and residual value rates. It is a screen image figure showing the tendency.
FIG. 16 shows a graph with the new vehicle price on one axis and the successful bid price on the other axis. On the graph, the actual data of the new vehicle price and the successful bid price for the resold vehicle is displayed, and the new car price and the successful bid price are displayed. The correlation with the price is displayed. The correlation between the new vehicle price and the successful bid price and the actual data that is the basis of the correlation are displayed in different colors for the 3-year lease and the 5-year lease. The winning bid price (sales amount) may be a residual value ratio obtained by dividing the selling price by the new vehicle price.
In the embodiment shown in FIG. 16, the correlation with the new car price that has already been empirically recognized from the sale data on the vehicles that have already been resold has been empirically recognized to have a large impact on the sale price. In addition, the objective sale price can be recognized for a vehicle that is scheduled for resale.

FIG. 17 shows a graph with the remaining distance ratio obtained by dividing the mileage by one axis and the sale price divided by the new vehicle price on the other axis, and the actual distance between the mileage and the remaining ratio for the resold vehicle is displayed on the graph. In addition to displaying data, the correlation between mileage and residual value rate is displayed. The correlation between the mileage and the residual value rate and the actual data on which the mileage is based are displayed in different colors for the 3-year lease and the 5-year lease. The residual value rate may be a successful bid price (sales amount).
In the embodiment shown in FIG. 17, the correlation with the travel distance that has already been empirically recognized from the sale data regarding the vehicles that have already been resold as having a large effect on the sale price is visually grasped. In addition, the objective sale price can be recognized for a vehicle that is scheduled for resale.

FIG. 18 shows a graph in which the lease period is one axis and the residual value ratio obtained by dividing the sale price by the new vehicle price is the other axis. On the graph, the mileage and the residual value ratio for the resale vehicle are shown. In addition to displaying data, the correlation between mileage and residual value rate is displayed. The residual value rate may be a successful bid price (sales amount).
In the embodiment shown in FIG. 18, the correlation with the lease period that has already been empirically recognized from the sale data relating to the vehicles that have already been resold has been empirically recognized as having a large impact on the sale price. In addition, the objective sale price can be recognized for a vehicle that is scheduled for resale.
3 to 8 already described in the analysis of the correlation, the same effects as described above can be obtained by displaying them as screen image diagrams for obtaining the overall tendency of the expected sale amount and the expected residual value rate.
In addition, about the travel distance demonstrated in the said Example, an actual use condition can be expressed further more accurately by making it an average travel distance in predetermined periods, such as a month and a year, for example besides a total travel distance.
Moreover, in the said Example, although demonstrated taking the vehicle as an example, it is applicable also to articles, such as a ship, a machine tool, equipment, and a personal computer other than a vehicle. In addition, the article in the present invention may be software such as a program, and is a concept including real estate such as a house or a building and equipment, not necessarily movable property.

The block diagram which shows the whole structure containing the vehicle resale price analysis system by one Example of this invention. Processing flow showing a method for acquiring a residual value calculation equation according to an embodiment of the present invention Scatter chart with X axis as new car price and Y axis as sale price Scatter chart with the X-axis as travel distance and Y-axis as the residual value ratio on sale according to this example Scatter chart with X-axis as new car price and Y-axis as average distance conversion sales amount Scatter chart with the X-axis as travel distance and the Y-axis as the average new car price conversion residual value ratio according to this embodiment Scatter plot with rank according to this embodiment as X axis and ARZ as Y axis Scatter plot with adjustment of normal equation in Fig. 7 Processing flow showing a method for obtaining a residual value calculation equation according to another embodiment of the present invention The graph which shows the heavy decision index at the time of selecting the number of elapsed months, displacement, new car price, and monthly mileage as an item by this example In the system according to an embodiment of the present invention, a screen image diagram for obtaining an expected sale amount and an expected residual value rate for a specific vehicle during a contract period or a specific vehicle at the time of a new contract In the system according to an embodiment of the present invention, a screen image diagram for predicting the residual value profit and loss based on the expected residual value for a specific vehicle during the contract period In the system according to an embodiment of the present invention, a screen image diagram for predicting the residual value profit / loss of each vehicle type for a vehicle in a contract period In the system according to an embodiment of the present invention, a screen image diagram for predicting the residual value profit and loss based on the expected residual value for a specific vehicle during the contract period In the system according to an embodiment of the present invention, a screen image diagram for predicting the residual value profit and loss based on the expected residual value for a specific vehicle during the contract period In the system according to an embodiment of the present invention, a screen image diagram showing trends in new vehicle prices and successful bid prices by lease period. In the system according to an embodiment of the present invention, a screen image diagram showing trends in mileage and residual value rate by lease organization. The screen image figure showing the tendency of a lease period and a residual value rate in the system by one example of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Resale support system 11 Support side vehicle resale system 12 Bid support system 13 Vehicle type database 14 Existing resale vehicle database 15 Expected sale price calculation system 16 Auction site system 20 User side system 21 User side vehicle resale price analysis system 22 Residual value setting system 23 Residual value simulation system 24 Residual value profit and loss analysis system 25 Asset evaluation system 26 Estimate system 27 Core system 28 Lease contract database 29 Exhibition support system

Claims (7)

  A residual value calculation method in a residual value calculation system for calculating a residual value of a vehicle,
  Vehicle sales data at the time of resale, vehicle sales data before resale obtained from the result of vehicle bidding, and vehicle usage status data that are determined to have a large effect on the sales price. Deriving correlation with actual usage data
  Based on the correlation, a remaining value calculation method of calculating the sale price data of the vehicle before resale from the sale price data of the vehicle at the time of resale and the actual use data of the vehicle at the time of resale.   Holds the residual value calculation formula for calculating the residual value of the vehicle generated for each attribute of the vehicle,
  Identify the attribute of the target vehicle for residual value calculation, and acquire the residual value calculation formula corresponding to the attribute of the target vehicle for residual value calculation from the retained residual value calculation formula,
  2. The residual value according to claim 1, wherein a residual value of the target vehicle of the residual value calculation is calculated by substituting a numerical value corresponding to the target vehicle of the residual value calculation into the acquired residual value calculation formula. Calculation method.   The numerical value corresponding to the target vehicle for calculating the residual value is a price, a travel distance, and an elapsed period since the registration of the new vehicle of the target vehicle for calculating the residual value. Residual value calculation method.   4. The residual value of the target vehicle for calculating the residual value is corrected based on at least one of the degree of popularity and the degree of wear of the target vehicle for calculating the residual value. The residual value calculation method as described.   The target vehicle for calculating the residual value is a new model,
  The residual value calculation method according to claim 2, wherein the residual value calculation formula is generated based on an existing vehicle.   A residual value setting simulation system for predicting the market price at an arbitrary point in time for the vehicle;
  A vehicle type database for storing data indicating vehicle attributes;
  An existing resale database for storing resale data relating to vehicles at the time of resale and pre-resale data relating to vehicles before resale obtained as a result of a vehicle bidding
With
  The data stored in the vehicle type database and the resale database is transmitted periodically or when data is updated from the vehicle type database and the resale database to the residual value setting simulation system. Vehicle resale price analysis system.   The residual value setting simulation system includes:
  Deriving the correlation between the existing resale data, the pre-resale data, and the actual use data that is determined to have a large effect on the sale amount, which is data related to the use state of the vehicle,
  7. The vehicle according to claim 6, wherein the vehicle sales amount data before resale is predicted from the vehicle sales amount data at the time of resale and the actual use data of the vehicle at the time of resale based on the correlation. Resale price analysis system.