Bankruptcy risk factor identification device and bankruptcy risk factor identification method

Abstract

The present invention provides a bankruptcy risk factor identification device and a bankruptcy risk factor identification method that identify bankruptcy risk factors based on the probability of bankruptcy calculated using stochastic differential equations. [Solution] The bankruptcy risk factor identification device according to the present invention comprises: a data storage unit that stores accounting data of a target company for a predetermined period provided by an accounting system; a variable parameter calculation unit that calculates the drift and volatility of the target company's assets and liabilities using the accounting data for the predetermined period; a bankruptcy probability calculation unit that calculates the bankruptcy probability by applying the calculated drift and volatility of the assets and liabilities to a stochastic differential equation that represents the target company's net assets in a stochastic process; a loop operation instruction unit that changes the value of the drift and / or volatility of any of the assets and liabilities applied to the stochastic differential equation by a predetermined step size and transmits it to the bankruptcy probability calculation unit, causing the bankruptcy probability calculation unit to repeatedly recalculate the bankruptcy probability; and a risk factor determination unit that identifies the drift and / or volatility value at a position where the bankruptcy probability exceeds a predetermined threshold in a probability distribution plotted against the changed drift and / or volatility value as a bankruptcy risk factor for the target company.

Description

[Technical Field]

[0001] The present invention relates to a bankruptcy risk factor identification device and a bankruptcy risk factor identification method, and more specifically, to a bankruptcy risk factor identification device and a bankruptcy risk factor identification method that identify bankruptcy risk factors using a stochastic differential equation that represents the financial condition of a company in the process of establishing it based on accounting data. [Background technology]

[0002] Corporate bankruptcy largely depends on management decisions and financial conditions. Traditionally, assessing bankruptcy risk has centered on the analysis of financial indicators and predictions based on analysts' empirical rules. Therefore, techniques have been proposed to predict the probability of a specific company's bankruptcy or business closure using statistical models.

[0003] For example, Patent Document 1 discloses a technique for generating a statistical model based on data of predetermined items selected based on expert knowledge from a large amount of company survey reports created by a credit rating agency, and then applying the data of a specific company for the same items to this model to calculate the probability that the company will cease operations within a predetermined period. Furthermore, Patent Document 2 discloses a technique for estimating the risk of a specific company ceasing operations after an arbitrary period of time by using a trained model generated by inputting model generation data related to registered companies into a machine learning algorithm that uses the survival time of a bankrupt company as the dependent variable and predetermined items (sales, number of employees, etc.) as independent variables.

[0004] However, these conventional technologies primarily focused on risk assessment (calculation of bankruptcy probability) from the perspective of stakeholders such as a specific company's business partners and financial institutions. They did not provide useful information for the management of the specific company involved, such as identifying factors that could affect their own bankruptcy and enabling appropriate decision-making for business improvement or bankruptcy avoidance. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2019-117443 [Patent Document 2] Japanese Patent Publication No. 2023-108896 [Overview of the project] [Problems that the invention aims to solve]

[0006] The present invention has been made in view of the above-mentioned prior art, and the object of the present invention is to provide a bankruptcy risk factor identification device and a bankruptcy risk factor identification method that automates the process of identifying bankruptcy risk factors based on the probability of bankruptcy calculated using a stochastic differential equation that represents the financial condition of a company as a stochastic process based on accounting data, and deriving the fluctuation trends of assets and liabilities that constitute bankruptcy risk. [Means for solving the problem]

[0007] An insolvency risk factor identification device according to one aspect of the present invention, made to achieve the above objective, is an information processing device that identifies insolvency risk factors based on the probability of insolvency calculated using a stochastic differential equation, the information processing device comprising: a data storage unit that stores accounting data of a target company for a predetermined period provided from an accounting system; a fluctuation parameter calculation unit that calculates the drift and volatility of the target company's assets and liabilities using the accounting data for the predetermined period; and a stochastic differential equation that expresses the calculated drift and volatility of the target company's net assets in a stochastic process. The system is characterized by comprising: a bankruptcy probability calculation unit that calculates the probability of bankruptcy by applying it to a differential equation; a loop operation instruction unit that repeatedly recalculates the bankruptcy probability by changing the drift and / or volatility value of any of the assets and liabilities applied to the stochastic differential equation by a predetermined step size; and a risk factor determination unit that identifies the drift and / or volatility value at a position where the bankruptcy probability exceeds a predetermined threshold in a probability distribution plotted against the changed drift and / or volatility value as a risk factor for the bankruptcy of the target company.

[0008] A method for identifying bankruptcy risk factors according to one aspect of the present invention, made to achieve the above objective, is a method for identifying bankruptcy risk factors by calculating the probability of bankruptcy using a stochastic differential equation with an information processing device, characterized in that the information processing device includes the steps of: acquiring accounting data for a predetermined period of a target company from an accounting system that provides accounting data for the target company; calculating the drift and volatility of the target company's assets and liabilities using the accounting data for the predetermined period; applying the calculated drift and volatility of the assets and liabilities to a stochastic differential equation representing the target company's net assets in a stochastic process to calculate the probability of bankruptcy; repeatedly recalculating the probability of bankruptcy by changing the value of the drift and / or volatility of any of the assets and liabilities applied to the stochastic differential equation by a predetermined step size; and identifying the drift and / or volatility at a position where the probability of bankruptcy exceeds a predetermined threshold in a probability distribution plotted against the changed drift and / or volatility values ​​as a bankruptcy risk factor for the target company. [Effects of the Invention]

[0009] According to the bankruptcy risk factor identification device and method of the present invention, a user can simulate the fluctuation trends (e.g., rate of change and magnitude of change) of assets and liabilities that could lead to the bankruptcy of a target company, using accounting data created by general accounting software, by setting various conditions. Furthermore, since the process of calculating the probability process and fluctuation trends leading to bankruptcy is automated, even those with little advanced expertise or programming experience can more accurately grasp their company's financial situation and factors that pose a bankruptcy risk with simple operations, thereby supporting appropriate decision-making to avoid bankruptcy. [Brief explanation of the drawing]

[0010] [Figure 1] This figure shows the overall configuration of a bankruptcy risk factor identification device according to one embodiment of the present invention. [Figure 2]It is a block diagram showing an example of the functional configuration of a bankruptcy risk factor identification device according to an embodiment of the present invention. [Figure 3] It is a flowchart for explaining the procedure of identifying bankruptcy risk factors by a bankruptcy risk factor identification device according to an embodiment of the present invention. [Figure 4] It is a diagram showing an example of the result of calculating the probability process of the net assets of a certain company by the bankruptcy risk factor identification device according to the present embodiment.

Embodiments for Carrying Out the Invention

[0011] Hereinafter, specific examples of the embodiments for carrying out the present invention will be described in detail with reference to the drawings.

[0012] FIG. 1 is a diagram showing the overall configuration of a bankruptcy risk factor identification device according to an embodiment of the present invention.

[0013] The bankruptcy risk factor identification device 10 according to an embodiment of the present invention shown in FIG. 1 is composed of a general computer (information processing device), has a central processing unit (processor) and ROM and RAM (not shown), etc., and executes an OS (Operating System) and a predetermined program. A control unit 11, a storage unit 12 composed of an SSD (Solid State Drive), an HDD (Hard Disk Drive), etc. for storing various programs and data, an input unit 13 for receiving instructions from the user and inputs of various data via various input means, and an input / output information and operation results are displayed on an image display means. A display unit 14, an output unit 15 for outputting data to a printer for writing on a paper medium and / or a data writing device for writing on a computer-readable recording medium, and a communication unit 16 for communicating with an external accounting system 20 and / or a client terminal 30 via various communication networks.

[0014] The input means connected to the input unit 13 may be, for example, a keyboard, mouse, or various data reading devices. The input unit 13 and the display unit 14 may be configured as an integrated input / output unit, such as a touch panel that integrates the input means and image display means. The communication unit 16 includes various communication modules for wired / wireless communication.

[0015] The external accounting system 20 is a computer system (server) configured to perform calculations related to a company's accounting operations, and the client terminal 30 is a computer terminal used by users of the bankruptcy risk factor identification device 10. The accounting system 20 is not limited to being composed of a single server, but may also be configured as a cloud system. Furthermore, the client terminal 30 may consist of a PC (Personal Computer), tablet PC, smartphone, etc.

[0016] Next, the functional configuration of the bankruptcy risk factor identification device 10 according to one embodiment of the present invention will be described.

[0017] Figure 2 is a block diagram showing an example of the functional configuration of a bankruptcy risk factor identification device according to one embodiment of the present invention.

[0018] In the bankruptcy risk factor identification device 10 shown in Figure 2, the control unit 11 reads predetermined programs and data and executes them in the central processing unit (processor) to realize multiple functional units and control the entire device including these functional units. The processor may include a GPU (Graphics Processing Unit) in addition to a CPU (Central Processing Unit).

[0019] The bankruptcy risk factor identification device 10 includes a data management unit 110, a variable parameter calculation unit 111, a bankruptcy probability calculation unit 112, a loop calculation instruction unit 113, and a risk factor identification unit 114, as functional units realized by having the processor of the control unit 11 execute a predetermined program (an application program for identifying bankruptcy risk factors).

[0020] The storage unit 12 is provided with a data storage unit 120, an analysis model storage unit 121, and a calculation result storage unit 122, which are storage areas corresponding to the content to be stored. The storage unit 12 is not limited to being composed of a single storage device, but may be composed of multiple storage devices connected to a network. That is, the storage unit 12 may be built into the computer that constitutes the bankruptcy risk factor identification device 10 or may be composed of an external data server connected to a network.

[0021] The following describes the functional components included in the bankruptcy risk factor identification device 10 shown in Figure 2.

[0022] The data management unit 110 assigns an identifying company ID to the accounting data and company information of companies subject to bankruptcy risk factor identification (hereinafter referred to as "target companies") for a predetermined period, which are received via the input unit 13 or the communication unit 16. It then stores (saves) the predetermined financial data included in the accounting data as time-series data, such as monthly or daily, for each account (data item) in the data storage unit 120 of the storage unit 12. The predetermined period can be one year or multiple years, with the accounting period as the unit.

[0023] The variable parameter calculation unit 111 uses the financial data (time-series data) corresponding to assets and liabilities from the financial data of the target company stored in the data storage unit 120 to calculate the drift and volatility of the assets and liabilities included in the mathematical model (in this embodiment, a model applying a structural approach) that calculates the probability of bankruptcy expressed by a stochastic differential equation, as described later. Here, the drift of assets is also called the average growth rate, and the drift of liabilities is also called the average growth rate. The volatility of assets and liabilities corresponds to their respective standard deviations. The variable parameter calculation unit 111 transmits the calculated drift and volatility of assets and liabilities to the bankruptcy probability calculation unit 112 and the loop calculation instruction unit 113, and also stores them for each target company in the analysis model storage unit 121 of the storage unit 12.

[0024] The financial data constituting the assets and liabilities of the mathematical model described above may include accounts (account titles) commonly found on a balance sheet. However, the present invention is not limited to this. In one embodiment of the present invention, the bankruptcy risk factor identification device 10 is configured to allow the user to set and change the accounts to be included in the assets and liabilities of the mathematical model, according to the size and form of the target company. The set accounts (data items) to be included in the assets and liabilities are stored in the analysis model storage unit 121 of the storage unit 12 for each target company and are referenced by the fluctuation parameter calculation unit 111 and the bankruptcy probability calculation unit 112. In the embodiment described later, for example, in the case where the target company is a small or medium-sized enterprise, the accounts to be included in the liabilities of the mathematical model are limited to interest-bearing debt.

[0025] The bankruptcy probability calculation unit 112 reads a mathematical model for calculating the bankruptcy probability of the target company from the analysis model storage unit 121 of the storage unit 12, and applies the drift and volatility of assets and liabilities calculated by the fluctuation parameter calculation unit 111 to the coefficients (drift coefficient and diffusion coefficient) of the stochastic differential equation that is the expression of the mathematical model, and calculates the bankruptcy probability. In addition, the bankruptcy probability calculation unit 112 applies the values ​​of the drift and / or volatility of assets and liabilities instructed by the loop calculation instruction unit 113 (described later) to the coefficients of the stochastic differential equation of the mathematical model and calculates the bankruptcy probability.

[0026] The loop calculation instruction unit 113 receives the drift and / or volatility values ​​of assets and liabilities from the variable parameter calculation unit 111, changes (increases or decreases) them by a predetermined step size, and transmits them to the bankruptcy probability calculation unit 112. The instruction unit 113 then instructs the bankruptcy probability calculation unit 112 to repeatedly perform the process of recalculating the bankruptcy probability by applying the changed drift and / or volatility values ​​of assets and liabilities to the coefficients of the stochastic differential equations of the mathematical model. The loop calculation instruction unit 113 also stores the bankruptcy probability calculated by the bankruptcy probability calculation unit 112 in the calculation result storage unit 122, associating it with the drift and volatility values ​​used in the calculation.

[0027] When the repeated recalculation process by the bankruptcy probability calculation unit 112 is completed, the risk factor identification unit 114 reads the recalculated bankruptcy probability results corresponding to the drift and / or volatility values ​​of the changed assets and / or liabilities from the calculation result storage unit 122. It generates a probability distribution graph plotting the read recalculated bankruptcy probability results against the drift and / or volatility values ​​of the changed assets and liabilities. From the generated probability distribution graph, it identifies the drift and / or volatility values ​​of the assets and liabilities that increase the probability of bankruptcy (the probability of bankruptcy exceeds a predetermined threshold) as bankruptcy risk factors for the target company. The generated probability distribution graph and the data of the identified bankruptcy risk factors are stored in the calculation result storage unit 122 of the storage unit 120. The threshold for the probability of bankruptcy at which a factor is determined to be a bankruptcy risk factor can be arbitrarily set by the user.

[0028] Here, we will describe in detail the mathematical model used to calculate the probability of bankruptcy performed by the bankruptcy probability calculation unit 112 of the bankruptcy risk factor identification device 10 according to one embodiment of the present invention. The mathematical model applied to this embodiment describes the financial state of a company as a stochastic process and adopts a so-called first passage time model, which defines default (i.e., bankruptcy) as the point at which the company's value (value of company assets) falls below a predetermined boundary (threshold). However, the present invention is not limited thereto.

[0029] The mathematical model used to calculate the probability of bankruptcy performed by the bankruptcy probability calculation unit 112 in this embodiment is: ·X t Total assets (all of a company's assets) ·L t Interest-bearing debt (corporate borrowings) ·D t Net assets (D t =X t -L t ) ·τ:(D t (The point at which it becomes 0) So,

[0030] (1) The probability process of total assets is represented by the following Equation 1.

[0031]

Equation

[0032] (2) The probability process of interest-bearing liabilities is represented by the following Equation 2.

[0033]

Equation

[0034] (3) The probability process of net assets is represented by the following Equation 3.

[0035]

Equation

[0036] Bankruptcy occurs at the time point τ when D t = 0.

[0037] Therefore, the probability of corporate bankruptcy is defined as the probability that the net assets D t reach zero, and based on the theory of first passage time, it can be approximated by the following Equation 4.

[0038]

Equation

[0039] Here, Φ(x) is the cumulative distribution function of the standard normal distribution, D0 = X 0- L0 is the current net assets, and σ D is the volatility of net assets.

[0040] Note that the cumulative distribution function Φ(x) of the standard normal distribution is generally represented by the following Equation 5.

[0041]

Equation

[0042] The bankruptcy probability calculation unit 112 of the bankruptcy risk factor identification device 10 according to this embodiment can calculate the bankruptcy probability of the target company by calculating a numerical solution to the above-mentioned formula 4 using a predetermined numerical solution method.

[0043] Furthermore, the method for calculating the probability of bankruptcy by the bankruptcy probability calculation unit 112 may be performed by a Monte Carlo simulation based on equation 3 (this will be explained in detail in the embodiment described later).

[0044] Based on the results of a series of calculations that identify bankruptcy risk factors, derived from the calculation of the probability of bankruptcy described above, the user can change, add, or remove financial data items to be included in the assets and liabilities of the pre-registered default mathematical model. The changes to the data items and / or the data items to be added or removed, entered by the user via the input unit 13, are stored in the analysis model storage unit 121 for each target company and are referenced by the variable parameter calculation unit 111 and the bankruptcy probability calculation unit 112.

[0045] Furthermore, the bankruptcy risk factor identification device 10 according to the present invention is not limited to the configuration consisting of a single computer as described above, but may be configured with multiple network-connected computers. Alternatively, the bankruptcy risk factor identification device 10 may be configured in a cloud-based manner, which performs bankruptcy risk factor identification processing in response to requests from external client terminals 30 and transmits the processing results to the client terminals 30.

[0046] The following describes a method for identifying a company's bankruptcy risk factors using a bankruptcy risk factor identification device according to one embodiment of the present invention, with reference to Figures 2 and 3.

[0047] Figure 3 is a flowchart illustrating the procedure for identifying bankruptcy risk factors using a bankruptcy risk factor identification device according to one embodiment of the present invention.

[0048] Referring to Figure 3, the control unit 11 of the bankruptcy risk factor identification device 10 accepts input of accounting data and company information of the target company via the input unit 13 after startup (S100). The input accounting data includes, for example, the financial data of the target company for the past year created using general accounting software, and each data is provided in CSV format. The company information includes the target company's company code, industry, trade name, representative, address, listing status, date of establishment, capital, number of employees, etc.

[0049] If the accounting data of the target company is created using accounting software implemented in a computer constituting the bankruptcy risk factor identification device 10, the user inputs a command via the input unit 13 to specify the bankruptcy risk factor identification device 10 as the output destination for the accounting data created by the accounting software. The control unit 11 transmits the accounting data output from the accounting software, along with pre-registered company information, to the data management unit 110 in accordance with the input command, and causes the data management unit 110 to execute the process of registering (storing) the data in a predetermined database pre-constructed in the data storage unit 120 of the storage unit 12 (step S110).

[0050] On the other hand, if the accounting data of the target company is provided from a network-connected accounting system 20 or client terminal 30, the user sends a command to the bankruptcy risk factor identification device 10 requesting the registration of accounting data and company information from the accounting system 20 or client terminal 30. The control unit 11 of the bankruptcy risk factor identification device 10 receives the accounting data and company information from the accounting system 20 or client terminal 30 via the communication unit 16 and causes the data management unit 110 to execute step S110, which registers the data in a predetermined database in the data storage unit 120 of the storage unit 12.

[0051] In step S110, which registers the received accounting data and company information in a predetermined database of the data storage unit 120, the data management unit 110 distributes and registers (stores) the monthly (or daily) financial data and company information included in the received accounting data into data items placed in the database tables. At this time, if data corresponding to a specific data item among the data items placed in the database tables is not included in the received accounting data, the data management unit 110 may perform a process to calculate and register the data value of the specific data item not included in the received accounting data using the source data of the relevant account included in the received accounting data.

[0052] Furthermore, users can provide accounting data and company information in a form stored on a computer-readable recording medium and input it via a reader for the recording medium connected to the input unit 13.

[0053] Next, when the control unit 11 of the bankruptcy risk factor identification device 10 receives an instruction from the user to start the bankruptcy risk factor identification process for the target company via the input unit 13, it activates the variable parameter calculation unit 111. The variable parameter calculation unit 111 reads the financial data (time-series data) for each item set as assets and liabilities in the mathematical model described above from the financial data included in the accounting data of the target company registered in step S110, from the database of the data storage unit 120, calculates a predicted value for each of the set financial data items using a predetermined time-series analysis model (e.g., the ARIMA model), and sums up the calculated predicted values ​​for each item to calculate the drift (μ) corresponding to assets (total assets) and liabilities (interest-bearing debt) in the stochastic differential equation (Equation 3) of the mathematical model. X , μ L The variable parameter calculation unit 111 calculates the standard deviation (volatility σ) of the predicted values ​​of the combined assets (total assets) and liabilities (interest-bearing debt). X , σ L ) are calculated respectively (step S120).

[0054] Furthermore, the method for calculating the coefficients (drift coefficient = drift and diffusion coefficient = volatility) for assets and liabilities set in the stochastic process of the mathematical model is not limited to the method described above. For example, maximum likelihood estimation may be used.

[0055] Subsequently, the variable parameter calculation unit 111 stores the calculated drift and volatility of assets and liabilities in the database of the target company in the analysis model storage unit 121, and also transmits them to the bankruptcy probability calculation unit 112 and the loop calculation instruction unit 113.

[0056] When the bankruptcy probability calculation unit 112 receives the calculated drift and volatility of the assets and liabilities for the target company from the fluctuation parameter calculation unit 111, it reads out the mathematical model for calculating the bankruptcy probability of the target company from the analysis model storage unit 121 of the storage unit 12 in order to calculate the bankruptcy probability, and applies the drift μ(μ) calculated in step S120 to each coefficient of the stochastic differential equation (equation 3) of the mathematical model. X , μ L (collectively referred to as) and volatility σ(σ) X , σ L The probability of bankruptcy is calculated by applying (substituting) the following (generally referred to as) the Monte Carlo method (Monte Carlo simulation) (step S130).

[0057] When the bankruptcy probability calculation unit 112 calculates the bankruptcy probability using the Monte Carlo method, the net assets D in the stochastic differential equation of formula 3 t The process (path) of change is calculated many times (for example, 100 to 10,000 times, but there is no particular limit), and net assets D t The probability of bankruptcy is calculated from the percentage of passes where the current value falls below 0 (zero) from t0 to a predetermined point in time (for example, one year later). The bankruptcy probability calculation unit 112 transmits the calculated bankruptcy probability, along with the drift and volatility values ​​used in the calculation, to the loop calculation instruction unit 113.

[0058] When the loop calculation instruction unit 113 receives the bankruptcy probability and the drift and volatility values ​​used in its calculation from the bankruptcy probability calculation unit 112, it associates the received bankruptcy probability with the drift and volatility values ​​used in its calculation and stores it in the calculation result storage unit 122 of the storage unit 120 for each target company. It also determines whether the drift and / or volatility values ​​used in the calculation have reached the upper or lower limit of a preset trial range, in other words, whether the calculation within the preset trial range has been completed (step S140).

[0059] Here, the pre-set trial range is the range in which the loop calculation instruction unit 113 changes the values ​​of drift and / or volatility (i.e., parameters) that cause the bankruptcy probability calculation unit 112 to perform the calculation of the bankruptcy probability. This range can be specified (input) by the user or a predetermined numerical range can be set in advance (an example is shown in Figure 4 below). The loop calculation instruction unit 113 uses the respective drift and volatility values ​​of assets and liabilities received from the variable parameter calculation unit 111 as initial values, and uses these initial values ​​as a reference (center) to change the drift and / or volatility values ​​of either the assets or liabilities within the set trial range by a predetermined step size. The user can specify which of the asset and liability drift and / or volatility values ​​to change.

[0060] When the loop calculation instruction unit 113 receives the initial bankruptcy probability calculated from the bankruptcy probability calculation unit 112 using the initial values ​​of the drift and volatility of the assets and liabilities, it stores the received bankruptcy probability in the calculation result storage unit 122, associating it with the drift and volatility values ​​used in the calculation. It also changes (increases or decreases) the drift and / or volatility value of either the assets or liabilities by a predetermined step size and sends it to the bankruptcy probability calculation unit 112 (step S145). The instruction unit 113 then instructs the bankruptcy probability calculation unit 112 to recalculate the bankruptcy probability by applying the changed drift and / or volatility value of either the assets or liabilities to the coefficients of the stochastic differential equation. The predetermined step size can be specified (input) by the user or set to a predetermined value in advance. Furthermore, the drift and volatility of the assets and liabilities to be changed (increased or decreased) can be changed independently, or they can be changed in any combination.

[0061] Subsequently, when the loop calculation instruction unit 113 receives the recalculated bankruptcy probability from the bankruptcy probability calculation unit 112, it stores the received bankruptcy probability in the calculation result storage unit 122, associating it with the drift and volatility values ​​used in the calculation, and determines whether the drift and / or volatility values ​​used in the calculation have reached the upper or lower limit of the preset trial range, that is, whether the calculation has been completed for each step within the preset trial range (step S140). If the upper or lower limit of the trial range has not been reached, the loop calculation instruction unit 113 sequentially changes (increases or decreases) the drift and / or volatility values ​​of either the asset or liability by a predetermined step and transmits this to the bankruptcy probability calculation unit 112 (step S145). It then returns to step S130 and instructs the bankruptcy probability calculation unit 112 to recalculate the bankruptcy probability by applying the changed drift and volatility values ​​of either the asset or liability to the coefficients of the stochastic differential equation.

[0062] Thereafter, the control unit 11 of the bankruptcy risk factor identification device 10 controls the bankruptcy probability calculation unit 112 and the loop calculation instruction unit 113 to repeat steps S130 to S140 to S145 until recalculations are completed for each step within the preset trial range. When the loop calculation instruction unit 113 completes the recalculation using the drift and / or volatility values ​​of either assets or liabilities for each step within the preset trial range, it sends a completion notification to the risk factor identification unit 114.

[0063] When the risk factor identification unit 114 receives notification from the loop calculation instruction unit 113 that the recalculation within the trial range is complete, it reads out the probability of bankruptcy calculated by applying the drift and / or volatility values ​​of any of the changed (increased or decreased) assets and liabilities to the coefficients of the stochastic differential equation of formula 3 from the calculation result storage unit 122, generates a probability distribution graph (2D or 3D) by plotting the read-out probability of bankruptcy against the drift and / or volatility values ​​used in its calculation, and identifies the drift μ and / or volatility σ values ​​at the position where the probability of bankruptcy increases (the probability of bankruptcy exceeds a predetermined threshold or forms a peak or local maximum) as the risk factors for the target company's bankruptcy (step S150). Note that there may be more than one local maximum in the distribution of the probability of bankruptcy.

[0064] Subsequently, the risk factor identification unit 114 plots the recalculated probability of bankruptcy for each target company in the calculation result storage unit 122, for each company. This data is called probability distribution data.

[0065] Figure 4 shows an example of the results of calculating (simulating) the stochastic process (path) of the net assets of a certain company (target company) using the bankruptcy risk factor identification device 10 according to this embodiment.

[0066] The example shown in Figure 4(a) shows that the bankruptcy risk factor identification device 10 according to this embodiment reads financial data for a period of any choice specified by the user, for example, the past year, from the database of the data storage unit 120, and performs the step (S120) of calculating the drift and volatility of assets and liabilities respectively using the financial data of the data items set for assets and liabilities in the mathematical model (stochastic differential equation) for calculating the bankruptcy probability of the target company from the financial data of the specified period. Then, the calculated drift and / or volatility of either the assets or liabilities is applied to the coefficients of the stochastic differential equation (Equation 3) that represents the net assets of the target company in a stochastic process, thereby calculating the net assets D t The process of calculating the transition (path) of net assets D is performed many times (for example, 100 to 10,000 times) (i.e., many paths are generated), and t In the step of calculating the probability of bankruptcy from the proportion of paths where the net asset D reaches zero (S130), i.e., in the process of performing a Monte Carlo simulation, the net asset D generated by numerous calculations is calculated. t This shows an example of part of the path.

[0067] In the bankruptcy risk factor identification method performed in the bankruptcy risk factor identification device 10 according to this embodiment, as shown in Figure 4(b), the loop calculation instruction unit 113 determines the financial status of the target company (net assets D t The values ​​(parameters) of the drift and / or volatility of either the asset or liability, which are coefficients of the stochastic differential equation representing the stochastic process, are changed (increased or decreased) from the initial values ​​(in Figure 4(b), drift = 0, volatility = 1) in predetermined increments and transmitted to the bankruptcy probability calculation unit 112, and the net asset D for each of the changed parameter values ​​is sent to the bankruptcy probability calculation unit 112. t The process involves calculating (generating) multiple paths (Figure 4(b) shows only one example of a path where the drift is changed to -0.3, 0, 0.1, and 0.3 while keeping the volatility unchanged). The process generates a large number of net asset D values ​​using the procedure described above. tBy performing a step for each parameter to calculate the probability of bankruptcy based on the proportion of paths that fall below 0 (zero) from the current time t0 to a predetermined time (for example, one year later), the probability of bankruptcy for multiple parameters (data points) within a trial range specified by the user or set in advance can be determined.

[0068] The control unit 11 of the bankruptcy risk factor identification device 10 according to this embodiment automatically executes the above-described simulation according to user input and controls the display unit 14 to display a graph of the stochastic process (path) of net assets exemplified in Figure 4 (selectable to display all paths at once or sequentially for each path) on the image display means so that the user can visually check or review the simulation results. This allows the user to easily grasp the stochastic process leading to bankruptcy.

[0069] Furthermore, in step S150 of the bankruptcy risk factor identification device 10 according to this embodiment, the control unit 11 controls the display unit 14 so that the risk factor identification unit 114 plots the calculated bankruptcy probability data against the drift and / or volatility values ​​of either the changed assets or liabilities, graphs (not shown), and displays it on the image display means in a predetermined display format. This allows the user to visually determine the degree of influence of the drift μ and volatility σ of assets and liabilities that constitute bankruptcy risk factors, including a wider range of μ and σ that includes their vicinity.

[0070] The bankruptcy risk factor identification device of the present invention allows users to obtain bankruptcy prediction results based on mathematical models without requiring advanced expertise. Furthermore, the bankruptcy risk factor identification device according to the present invention can also reflect (modify) specific accounts or periods from the financial data included in the received accounting data into the mathematical model. For example, a user can adjust (increase or decrease) the parameters (μ, σ) of any one of the accounts included in the assets and liabilities of a pre-registered default mathematical model to investigate the impact of that account on the probability of bankruptcy, or change the period of the stochastic process to see short-term or long-term trends. This allows users to understand their company's financial situation and the factors that pose a bankruptcy risk in more detail. In this case, the bankruptcy risk factor identification device 10 may further include a configuration that accepts user input to set the parameters of the account to be changed and the target period, and reflects this in the mathematical model.

[0071] As described above, the bankruptcy risk factor identification device of the present invention can automate the process of simulating fluctuation trends in assets and liabilities that could lead to the bankruptcy risk of a target company, using accounting data of the target company that has been prepared in advance by general accounting software. Therefore, it is possible to reduce the effort and cost required to aggregate the large amount of statistical data that was previously necessary for identifying bankruptcy risk factors.

[0072] Furthermore, since users can change the fluctuation trends of assets and liabilities themselves to identify risk factors for bankruptcy, managers can proactively simulate measures to avoid bankruptcy, thereby supporting appropriate decision-making.

[0073] Although embodiments of the present invention have been described in detail above with reference to the drawings, the present invention is not limited to the embodiments described above, and can be modified and implemented in various ways without departing from the technical scope of the present invention. [Explanation of Symbols]

[0074] 10. Bankruptcy Risk Factor Identification Device 11 Control Unit 12 Storage section 13 Input section 14 Display section 15 Output section 16 Communications Department 20 Accounting Systems 30 client terminals 110 Data Management Department 111 Variable parameter calculation unit 112 Bankruptcy Probability Calculation Unit 113 Loop Calculation Instruction Unit 114 Risk Factor Identification Department 120 Data Storage Unit 121 Analysis Model Memory Unit 122 Operation result storage section

Claims

1. An information processing device that identifies bankruptcy risk factors based on the probability of bankruptcy calculated using a stochastic differential equation, The aforementioned information processing device is A data storage unit that stores accounting data for a specified period of the target company provided by the accounting system, A fluctuation parameter calculation unit that calculates the drift and volatility of the assets and liabilities of the target company using accounting data for the predetermined period, A bankruptcy probability calculation unit calculates the probability of bankruptcy by applying the respective drifts and volatility of the assets and liabilities calculated above to a stochastic differential equation that represents the target company's net assets in a stochastic process, A loop operation instruction unit that causes the bankruptcy probability calculation unit to repeatedly recalculate the bankruptcy probability by changing the drift and / or volatility value of any of the assets and liabilities applied to the stochastic differential equation by a predetermined step size, A bankruptcy risk factor identification device comprising: a risk factor determination unit that identifies the drift and / or volatility values ​​at positions where the bankruptcy probability exceeds a predetermined threshold in a probability distribution plotted against the changed drift and / or volatility values ​​as bankruptcy risk factors for the target company.

2. The aforementioned assets and liabilities are defined as total assets and interest-bearing debt, and the fluctuation parameter calculation unit uses the accounting data to calculate the drift and volatility of the target company's total assets and interest-bearing debt, respectively. The bankruptcy risk factor identification device according to claim 1, characterized in that the bankruptcy probability calculation unit applies the calculated drift and volatility of total assets and interest-bearing debt to the stochastic differential equation.

3. The bankruptcy risk factor identification device according to claim 2, characterized in that the bankruptcy probability calculation unit determines the point in time when net assets become zero in the numerical solution of the stochastic differential equation as a bankruptcy and calculates the bankruptcy probability.

4. A method for identifying bankruptcy risk factors based on the probability of bankruptcy calculated using a stochastic differential equation by an information processing device, The aforementioned information processing device The steps include: obtaining accounting data for a specified period of the target company from an accounting system that provides accounting data for the target company; A step of calculating the drift and volatility of the assets and liabilities of the target company using accounting data for the predetermined period, The steps include: applying the calculated drift and volatility of the assets and liabilities to a stochastic differential equation representing the net assets of the target company in a stochastic process to calculate the probability of bankruptcy; A step of repeatedly recalculating the probability of bankruptcy by changing the drift and / or volatility values ​​of any of the assets and liabilities applied to the aforementioned stochastic differential equation in predetermined increments, A method for identifying bankruptcy risk factors, characterized by comprising the step of identifying the drift and / or volatility values ​​at positions where the bankruptcy probability exceeds a predetermined threshold in a probability distribution obtained by plotting the recalculated bankruptcy probability against the changed drift and / or volatility values ​​as bankruptcy risk factors for the target company.

5. Total assets and interest-bearing debt are defined for the aforementioned assets and liabilities. The step of calculating the drift and volatility includes the step of calculating the drift and volatility of the total assets and interest-bearing debt of the subject company, respectively, using the accounting data. The method for identifying bankruptcy risk factors according to claim 4, characterized in that the step of calculating the probability of bankruptcy includes a step of applying the calculated drift and volatility of total assets and interest-bearing debt to the stochastic differential equation.

6. The method for identifying bankruptcy risk factors according to claim 5, characterized in that the step of calculating the bankruptcy probability is to determine the point in time when net assets become zero in the numerical solution of the stochastic differential equation as bankruptcy and calculate the bankruptcy probability.

Images