RU126169U1 - AUTOMATED INFORMATION ANALYTICAL CREDIT PORTFOLIO MANAGEMENT SYSTEM - Google Patents

Abstract

An automated information and analytical system for managing loan portfolios containing at least one operator’s workplace, including a data input device, a device for receiving data on financial instruments, data on the main characteristics of financial instruments, data on transactions with financial instruments, data on historical values risk factors, intermediate data storage and processing devices and data output device, while the operator’s workplace is connected by communication lines with rverom comprising means for generating primary transition matrices and loading behavior of the main base loan portfolio; a means of loading historical and generating scenario quality characteristics and volumes of generations of a loan portfolio, as well as loading pricing policies; a means of loading historical macro factor values, as well as generating scenario macro factor values; means for calculating the qualitative characteristics and macro factor, for calculating the scenario characteristics of the loan portfolio; a means of calculating the cost of funding, interest payments, losses resulting from the realization of market risk, as well as a means of generating reports using reporting algorithms based on data calculated by the aforementioned system server tools, moreover, managing means of generating primary transition matrices and loading the main loan portfolio behavior database, calculation funding costs, interest payments, losses resulting from the realization of market risk, loading of historical values of the macro factor, and also the formation of the stage

Description

Technical field

The utility model relates to computing tools designed to solve special problems, namely, to carry out predictive assessments of credit, market and strategic risks, modeling scenarios of managing loan portfolios (business scenarios), modeling macroeconomic scenarios and choosing optimal management decisions to reduce these risks. The utility model can be used to manage loan portfolios of banking and other financial institutions.

State of the art

Various systems are known for forming and managing loan portfolios. Putting these systems into practice requires not only specialized knowledge, but is almost always an art, so an important task is to assess the quality of the existing portfolio - to control the perfect managerial actions. As a rule, the final result of assessing the quality of the loan portfolio is taken as a simple assessment based on statistics on the occurrence of individual loans in the loan portfolio in arrears for a certain period. However, assessing the quality of the loan portfolio only by observing statistics on delinquency cannot be considered consistent, since it does not take into account external factors affecting the loan portfolio and cannot guarantee the same result in the future. Assessment of the effectiveness of the loan portfolio should be comprehensive and based not only on the absolute figures of losses received in a particular period of time, but also on macroeconomic data, on the assessment of the strength of the impact of the economic situation on the loan portfolio.

Closest to the claimed utility model is a system for managing a mortgage portfolio and a portfolio of consumer loans (Patents No. 86249775 B1, 06/19/2001; No. 57020631 B2, 03/28/2006). The mortgage and consumer loan portfolio management system was developed in the form of an analytical tool to improve the analysis of portfolio behavior in the past and forecasting its future behavior. The system considers portfolios at the vintage level by the time of issuing loans. Thanks to this grouping, it is possible to compare individual vintages. The early warning component of the system allows you to predict the expected level of delay in the portfolio for a certain time interval. The matrix component combines vintage analysis with the early warning component and predicts the level of delay in the loan portfolio at a certain point in the future. The results of the analysis are graphically displayed and automatically give an answer to the question of the advisability of investing in various loan portfolios.

The main goal of the system, considered as an analogue, is to provide assistance in understanding the behavior of the loan portfolio, to create an automated and dynamic system that is able to predict its future behavior by the portfolio's behavior and can also help in making decisions at the stage of the loan application.

However, considered as an analogue, the system is intended mainly for forecasting, not management, and in this system there are no means to choose the option of managing loan portfolios.

Thus, there is a need to expand the arsenal of technical management tools in the loan portfolio management system.

Utility Model Essence

The objective of the utility model is to expand the arsenal of technical management tools in the loan portfolio management system.

To solve the problem according to the utility model, an automated information and analytical system for managing loan portfolios has been implemented, which contains at least one operator’s workplace, including a data input device, a device for receiving data on financial instruments, data on the main characteristics of financial instruments, data on operations with financial instruments, data on historical values from |) risk actors, intermediate storage and data processing devices and devices data output, while the operator’s workstation is connected by communication lines to a server containing a means of generating primary transition matrices and loading the main behavior portfolio of the loan portfolio; a means of loading historical and generating scenario quality characteristics and volumes of generations of a loan portfolio, as well as loading pricing policies;

a means of loading historical macro factor values, as well as generating scenario macro factor values; means for calculating the qualitative characteristics and macro factor, for calculating the scenario characteristics of the loan portfolio; a means of calculating the cost of funding, interest payments, losses resulting from the realization of market risk, as well as a means of generating reports using reporting algorithms based on data calculated by the aforementioned system server tools, moreover, managing means of generating primary transition matrices and loading the main loan portfolio behavior database, calculation funding costs, interest payments, losses resulting from the realization of market risk, loading of historical values of the macro factor, and Also, the formation of scenario values of the macro factor, for calculating the qualitative characteristics and the macro factor, for calculating the scenario characteristics of the loan portfolio, loading historical and forming the scenario quality characteristics and generation volumes of the loan portfolio, as well as downloading price policies, is carried out from the operator’s workstation through the functional connection of the intermediate storage device data with computer subsystems. At the same time, the operator’s workplace is made possible, on the basis of reports received at the reporting stage, to choose a management option that is the most appropriate for the restrictions imposed and has the lowest risk values.

Brief Description of the Drawings

The utility model is illustrated by drawings, which show a General block diagram of a computer system for a preferred embodiment of the claimed utility model (Figure 1), a block diagram of the components of a server of a computer system (Figure 2), a block diagram of the subsystems of a computer system (Figure 3 ) and an auxiliary block diagram of the subsystems of the computer system (Figa).

Utility Model Disclosure

The information and analytical system for managing loan portfolios can be implemented on the basis of various computer systems, however, it is preferable to use a system in the form of a well-known computer network of the "client / server" type, including a server connected to the workstations (client computers) of the system operators and data sources with using various types of communication lines, for example fiber optic, telephone, radio, satellite communications, etc.

At the same time, a standard IBM-compatible personal computer or laptop computer using the well-known Microsoft Windows operating system or its other equivalent can be used as a workstation for the system operator. The system operator’s workstation contains a data input device (keyboard, mouse, touch screen, etc.), a data receiving device (memory on hard disks, a CD-ROM drive or a floppy disk drive), intermediate data storage and processing devices made in the form of random access memory of the client computer, a microprocessor, as well as data output devices (eg, display, printer).

The system comprises a server 1 connected to workstations (client computers) 2 of the system operators and data sources 3, various communication lines 4. The workstation 2 contains a data input device 5, a data receiving device 6, an intermediate data storage device 7, a microprocessor 8, and also data output devices 9. Server 1 in a preferred embodiment of the system comprises a microprocessor 10, random access memory 11 and storage media 12. Server 1 is a means of accommodating computer subsystems.

Sources of data 3 can be information systems implemented in the form of databases MS SQL Server, Oracle, SAS, Teradata, ODBC sources or the like, from which data are transmitted via communication lines, for example, fiber optic, telephone, satellite or other.

Server 1 contains:

means 13 for generating primary transition matrices and loading the main behavior base of the loan portfolio, consisting of a resource 14, a microprocessor 10, cells 15 of the RAM 11, segment 16 of the server’s hard disk 12;

means 17 for loading historical and generating scenario quality characteristics and volumes of generations of a loan portfolio, as well as downloading price policies, consisting of a resource 18, a microprocessor 10, cells 19 of a random access memory 11, a segment 20 of a server’s hard disk 12;

means 21 for loading historical values of the macro factor, as well as forming scenario values of the macro factor, consisting of a resource 22, a microprocessor 10, cells 23 of a random access memory 11, a segment 24 of a server’s hard disk 12;

means 25 for calculating the quality characteristics and macro factor, for calculating the scenario characteristics of the loan portfolio, consisting of a resource 26, a microprocessor 10, cells 27 of the RAM 11, segment 28 of the hard disk 12 of the server;

means 29 for calculating the cost of funding, interest payments, losses resulting from the realization of market risk, consisting of a resource 30, a microprocessor 10, cells 31 of RAM 11, segment 32 of the hard disk 12 of the server.

All means 13, 17, 21, 25, 29 contain modules for generating intermediate reports and data output, and data output is carried out by means of functional communication with the device 7 for intermediate data storage.

Server 1 also contains means 33 for generating reports by means of reporting algorithms based on data calculated by other means of the system, consisting of a resource 34, microprocessor 10, cells 35 of RAM 11, segment 36 of the server’s hard disk 12.

Means 13, 17 and 21 are associated with means 25 by organizing functional relationships between the memory cells 15 and 27, 19 and 27, 23 and 27, and means 25 and 29 are connected to each other by organizing functional communications between the operational cells 27 and 31. In turn, the functional connection of the cells 15 and 35, 19 and 35, 23 and 35, 27 and 35, 31 and 35 provides a serial connection of the means 13, 19, 23, 25 and 29 with the means 33. The output of the means 33 through the intermediate device storing and processing data 7 is connected to a data output device 9 of the system.

During the operation of the system, the system operator accesses from the workstation 2 to the server 1 means by means of a single graphical user interface and functional connection of the intermediate data storage device 7 with all computer subsystems.

On each of the aforementioned tools 13, 17, 21, 25, 29 of server 1, one computer subsystem is implemented, which is a collection of data banks designed to store data, and hardware and / or software and hardware, and / or software modules designed to organization of functional relationships between subsystems, the implementation of external data input into the system, data conversion and data transfer to other subsystems. A data bank is a set of ordered data located on physical data carriers — electromagnetic, optical, or any other media. Databanks belonging to different subsystems can be stored both separately in various databases, and jointly in one common database.

The tool 13 of the server 1 is designed to implement the subsystem of loading and storing data on the historical behavior of the loan portfolio, the formation of primary transition matrices (templates) (37).

The tool 17 is intended for the implementation of the subsystem of loading, formation and storage of qualitative characteristics of the loan portfolio (38).

The tool 21 is designed to implement the subsystem of loading historical data of macroeconomic indicators and their forecasts, the formation, modeling and storage of scenarios of macroeconomic indicators and macro-factor (39).

The tool 25 is designed to implement the subsystem settings algorithm for calculating the scenario characteristics of the portfolio, the calculation of these characteristics (40).

Tool 29 is intended for the implementation of the subsystem for calculating the cost of funding, interest payments and calculating market risk, and calculating the probable losses in the implementation of adverse scenarios (41).

The reporting module (42) is implemented on the tool 33.

As shown in FIG. 3, the subsystem (37) includes:

module (43) for loading and storing data on the historical behavior of the loan portfolio;

a module (44) for the automated formation of primary transition matrices (templates);

module (45), which implements an interface through which the formation and storage of primary transition matrices is directly carried out, the primary transition matrices calculated by module (44) or other systems are loaded;

module (46) contains the final versions of primary matrices (templates) intended for export to the portfolio characteristics calculation subsystem (40) and to the report preparation module (42).

Subsystem (38) includes:

a module (47) for loading and 4irming quality characteristics by generation (vintage) on loans already issued, through which qualitative characteristics are directly downloaded and generated (LTS, PD, TPD, 90 + @ 12MOV, 90 + @ 24MOV), as well as scenario characteristics are set future generations;

module (48) for exchanging scenario characteristics with the subsystem for calculating portfolio characteristics (40) and for export to the report preparation module (42).

Subsystem (39) includes:

module (49) for loading historical trends of macroeconomic indicators and their forecasts;

module (50) for the formation of macroeconomic scenarios and macro-factor (macro-factor is an internal variable of the system), as well as for modeling the stochastic process of macro-factor behavior;

module (51) for creating a matrix of sensitivity to macroeconomic influences (environmental influences), loading and exchanging historical and scenario values of the macro factor, as well as for exchanging macro factor values with the subsystem for calculating portfolio characteristics (40) and for exporting to the report preparation module (42).

Subsystem (40) includes:

a module (52) for generating a functional for minimizing errors;

a module (53) for calculating historical values of the qualitative characteristics of the loan portfolio and macro factor;

the module (54) for calculating the scenario characteristics of the loan portfolio includes a means of exporting data to the market risk calculation subsystem (41) and to the report preparation module (42).

Subsystem (41) includes:

module (55) for calculating and storing data on payments on the principal debt from new issues (details are provided for various types of loans, including the period for which loans are issued in the case of consumer loans or mortgages);

module (56) for loading and storing historical interest rates for different periods for various financial instruments available for a financial institution;

module (57) for calculating and storing data on the cost of funding for new grants for various funding scenarios (short-term borrowing, match-funding, ...);

module (58) for calculating and storing interest payment data for various funding scenarios;

module (59) for calculating and storing data on payments on the principal debt from the current portfolio (detailing is carried out for various types of loans, including the period for which loans are issued in the case of consumer loans or mortgages);

module (60) for calculating the current cost of debt servicing for various funding scenarios;

a module (61) for generating interest rate behavior scenarios;

a module (62) for generating stochastic time series of interest rate behavior;

module (63) for calculating funding costs, interest payments and calculating market risk for various scenarios of interest rate behavior and for various funding scenarios.

Modules 55, 57-61, and 63 contain a means of exporting data to the reporting module (42).

The action of the information-analytical system is as follows. The subsystem for loading and storing data on the historical behavior of the loan portfolio, the formation of primary transition matrices (37) downloads up-to-date data on the distribution of balances (pieces) by risk class, the transitions of balances (pieces) from the risk class to the risk class for each generation of loans , for each type of loan (for example, the term for issuing) loans. Data is downloaded for the entire life of the loan portfolio with monthly discretization (or with another discretization at the choice of the analytic operator). According to the base:

tenor opendate viewdate riskclass debtrur rcold contract debt perc mob irp 24 2007.2007 10.2008 0 915175 0 824 1231424 0.743 19 32351 24 2007.2007 10.2008 one 96220 0 68 1231424 0,078 19 0 24 2007.2007 10.2008 0 54501 one 41 218172 0.249 19 10328 24 2007.2007 10.2008 one 57926 one 40 218172 0.265 19 5181 24 2007.2007 10.2008 2 68170 one 36 218172 0.312 19 0

tenor - type of loan, loan term;

opendate - month of issuing loans;

viewdate - month of observation;

riskclass - risk class into which the balance is transferred for the observed month;

debtrur - balance that transfers from a risk class to a risk class;

rc_old - the risk class from which the balance is transferred for the observed month;

contract - the number of contracts that transfer from the risk class to the risk class;

debt - the initial balance in the risk class from which the transition is made;

prec is the percentage of transition;

mob - age of loan generation in months;

irp - the amount of interest payments made in the observed month for the corresponding group.

By processing data using the appropriate modules in the subsystem for loading and storing data on the historical behavior of the loan portfolio, generating primary transition matrices (37), data on the behavior of the loan portfolio are generated and primary transition matrices are generated, which are then transferred to the tuning subsystem of the algorithm for calculating the scenario characteristics of the portfolio and calculation of these characteristics (40).

At this stage, the primary transition matrices are controlled by plotting the probability functions of the transition from one risk class to another risk class depending on the age of the loan generation (44, 45). At this stage, it may be possible to adjust these functions by the system operator, as well as the ability to load primary transition matrices from other systems. Prepared primary transition matrices are stored (46) and transferred to the calculation subsystem (40), as well as to the report generation module (42).

Before starting the data processing procedure, the error minimization functional is set (52). Then, by processing the data using module (53), received from subsystem (37) into subsystem (40), the historical qualitative characteristics of the loan portfolio and macro factor are calculated. Estimated data on the qualitative characteristics of the loan portfolio enter the subsystem (38). The calculated data on the macro factor value enter the subsystem (39). At this stage, the actual and estimated characteristics of the loan portfolio are compared. If the deviation of the calculated characteristics from the actual ones is significant, then the fitting algorithm is changed, the functional and sequence of calculations are changed (52). If the deviation of the calculated from the actual characteristics of the loan portfolio is satisfactory, the final data of the qualitative characteristics of the loan portfolio and macro factor are transmitted to subsystems (38) and (39).

Subsystem (38) downloads price policies and data on the quality of loans from other systems, as well as data on the quality of loans from the subsystem for calculating the characteristics of the loan portfolio (40). At this stage, business scenarios are being formed, for each scenario, for each generation of future loans, the volume of disbursements, the quality and the structure of the terms are set. Scripts are transferred to the subsystem (40) and to the report preparation module (42).

In subsystem (39), macroeconomic indicators and their forecasts are loaded, as well as macro factors are loaded from the subsystem for calculating the characteristics of the loan portfolio (40). At this stage, the formation of macroeconomic scenarios and macro factor. Scripts are transferred to the subsystem (40) and to the report preparation module (42).

At the stage of calculating the scenario characteristics of the portfolio, a planning horizon is set, calculation procedures are selected from the general list (calculations for losses, write-offs, calculation of reserves, calculation of duration, calculation of interest payments and payments on principal and many other procedures) and the calculation process starts. The calculation results are transferred to the subsystem for calculating the cost of funding, interest payments and calculation of market risk, the calculation of probable losses in the implementation of adverse scenarios (41) and the reporting module (42).

At the stage of calculating market risk, interest rate data is downloaded. Estimated data on payments on the principal debt come from subsystem (40). The calculation of data on payments on the principal debt from new issues, on the current portfolio, on the cost of funding (for various scenarios of interest rate movements), on interest payments, and on the calculation of expected and unforeseen losses due to market risk is carried out. The calculated data are transmitted to the reporting module (42).

When transferring data to the report preparation module (42), the data is stored in the 36 segment of the hard disk 12 of the server 1. The data stored in the 36 segment of the hard disk is loaded into the memory cells 35 of the server 11 11. The report preparation module provides a whole range of reports for credit management , market and strategic risks (financial plans for income and expenses, credit losses, reserves, market losses, capital adequacy, anticipated and unforeseen losses, etc.). The system operator is provided with a wide range of options for choosing the method of construction and the data set used to build reports. To reflect the results of the system, all available and known means of display can be used. These may include tables, graphs, light and sound signals, software, service, help or other tips, etc.

Based on the reports received at the reporting stage, a management option is selected that is the most appropriate for the restrictions imposed and has the lowest risk values, i.e. choosing the best loan portfolio management option.

After that, data on the selected management option is stored, a long-term work plan of such departments of the financial and credit organization as risk department, financial department, treasury, marketing department, pricing department, and sales department is formed (adjusted). By agreement of the authorized person of the financial and credit organization, the corresponding updated work plan shall enter into force and become available to users working in these units.

When using the proposed automated information and analytical system, the efficiency of managing the loan portfolio increases and the level of risks in managing liabilities and assets decreases.

The system provides the opportunity to study loan portfolios of third-party organizations and the ability to build specialized reports for these organizations in order to develop professional recommendations for optimizing the management of loan portfolios.

The system provides for the possibility of using it by creating specialized reports for research purposes to study the laws of behavior of the loan portfolio, study the impact of the crisis on the loan portfolio, study the strength of the crisis and the geography of the crisis. For example, in the crisis of 2008-2009, the crisis hit mainly in regions with weak diversification of the economy (Chelyabinsk, Rostov-on-Don), while Moscow and St. Petersburg were practically outside the crisis impact zone.

Claims (1)

An automated information and analytical system for managing loan portfolios containing at least one operator’s workplace, including a data input device, a device for receiving data on financial instruments, data on the main characteristics of financial instruments, data on transactions with financial instruments, data on historical values risk factors, intermediate data storage and processing devices and data output device, while the operator’s workstation is connected by communication lines to a server containing means for generating primary transition matrices and loading the main behavior base of the loan portfolio; a means of loading historical and generating scenario quality characteristics and volumes of generations of a loan portfolio, as well as loading pricing policies; a means of loading historical macro factor values, as well as generating scenario macro factor values; means for calculating the qualitative characteristics and macro factor, for calculating the scenario characteristics of the loan portfolio; a means of calculating the cost of funding, interest payments, losses resulting from the realization of market risk, as well as a means of generating reports using reporting algorithms based on data calculated by the aforementioned system server tools, moreover, managing means of generating primary transition matrices and loading the main loan portfolio behavior database, calculation funding costs, interest payments, losses resulting from the realization of market risk, loading of historical values of the macro factor, and Also, the formation of scenario values of the macro factor for calculating the qualitative characteristics and the macro factor for calculating the scenario characteristics of the loan portfolio, loading of historical and formation of scenario quality characteristics and volumes of generations of the loan portfolio, as well as downloading of price policies is carried out from the operator’s workstation by means of functional connection of the intermediate data storage device with computer subsystems; wherein the operator’s workstation is able, based on the reports received at the reporting stage, to select a management option that is the most appropriate for the restrictions imposed and has the lowest risk values.

Images