MXPA02005432A - System and method for valuing loan portfolios using fuzzy clustering. - Google Patents

Abstract

L invention concerne un systeme et un procede permettant d evaluer des portefeuilles de prets au moyen d une logique floue. Ce systeme comprend une logique d acquisition d actifs qui acquiert une pluralite d elements d actif et attribue une pluralite de variables a cette pluralite d elements d actif dans une base de donnees d actif. Une logique de segmentation examine cette pluralite d elements d actif et de variables a l aide d un modele. Un logique de classification floue calcule ensuite la valeur de cette pluralite d elements d actif et de variables et une logique d analyse de rentabilite calcule la rentabilite de cette pluralite d elements d actif. Ce procede permet l evaluation des portefeuilles de prets au moyen d une logique de classification. Ce procede consiste a acquerir une pluralite d elements d actif et a appliquer une pluralite de variables a ces elements d actif. Cette pluralite d elements d actif et de variables est examinee a l aide d un modele et une logique floue est utilisee pour calculer la valeur de la pluralite d elements d actif et de variables. La rentabilite de la pluralite d element d actifs est ensuite calculee.

Description

SYSTEM AND METHOD FOR VALUING LOAN PORTFOLIOS USING INDISTINTABLE GROUPING CROSS REFERENCE TO RELATED REQUESTS This application claims the benefit of the Provisional Application of E.U.A. Series No. 60 / 168,499 filed on December 2, 1999, entitled "System and Method for Valuing Commercial Loans Using Indistinct Grouping and Knowledge of Engineering for Commercial Financing" (System and Method for Valuing Commercial Loans Using Fuzzy Clustering and Knowledge Engieneering for GECS Commercial Finance), which is incorporated herein by reference in its entirety.

BACKGROUND OF THE INVENTION This invention relates to the valuation of loan portfolios, and more specifically, describes a system and method for the valuation of loan portfolios using grouping logic. Generally, the valuation process involves the determination of the value of the income / cash flow assets.

It is a relatively simple process that can be applied to expected benefit streams of bonds, stocks, property leases, oil wells, and loans. The valuation process is carried out in order to determine the assets of "assigned value", at a point of time. The time value of the concept of money and the concept of return risk are key elements in the valuation process. In a very simple way, the valuation of any asset is equal to the net present value of all expected future benefits, which are commonly measured in terms of cash flow. Currently, it is difficult to estimate the profitability valuation of a portfolio of assets. Generally, an insurer must go through the loans in a portfolio one by one. This is very tedious and time consuming.

COMPENDIUM OF THE INVENTION This invention describes a system and method for the valuation of loan portfolios using indistinct logic. Briefly described, in architecture, the system can be implemented as follows. An asset acquisition logic obtains a plurality of assets and attaches a plurality of variables to the plurality of assets in an asset database. Segmentation logic examines the plurality of assets and variables with a model. An indistinct grouping logic calculates a value of the plurality of assets and variables, and a logic of profitability analysis calculates the profitability of the plurality of assets.

^^ M-.a¿fa¿t, ta.te, .J ". ^, ^. J | m < ata 'i fct This description can also be seen as providing a method for the valuation of loan portfolio using indistinct logic. In this regard, the method can be broadly summarized through the following steps: (1) acquire a plurality of assets and append a plurality of asset variables; (2) examine the plurality of assets and variables with a model; (3) use indistinct grouping to calculate the value of the plurality of assets and variables; and (4) calculate the profitability of the plurality of assets. These assets can be, but are not limited to, loan portfolios, bonds, stocks, property leases, etc.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram illustrating an example of asset valuation using the indistinct grouping system with the loan portfolio example. Figure 2 is a block diagram illustrating the valuation of the assets using the indistinct grouping, seated within a computer readable medium in a computer system. Figure 3 is a block diagram illustrating an example of the process flow of the system and method for the valuation of the assets using indistinct grouping. Figure 4 is a flow diagram illustrating the process flow of the system and method for asset valuation using . Mk m¿! Á > ¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡& Figure 5 is a block diagram of an example illustrating the different types of data acquired during the data acquisition process, as shown in Figures 2, 3 and 4. Figure 6 is a table of examples of critical variables and coding schemes used in the hierarchical segmentation process, as shown in Figures 3 and 4. Figure 7 is a block diagram of an example, illustrating the use of segmentation modeling for the loan portfolio example, as shown in Figures 2, 3 and 4 above. Figure 8 is an example of a flow diagram of the indistinct grouping process used in the asset valuation method and method of the present invention, for the example of loan portfolios, as shown in Figures 2, 3 and 4 Figure 9 is a flowchart of an example of the process calculating the indistinct grouping means in the system and method for valuation of assets of the present invention, for loan examples and a portfolio as shown in Figures 2. , 3, 4, and 8. Figure 10 is a flowchart of an example of the process of reviewing financing in the system and method for valuation of assets of the present invention, for the example of loans in a portfolio, as shown in Figures 2, 3, and 4 Figure 11A is a diagram illustrating an example of the variation between groups of six groups generated in the dendrogram, as shown in Figure 11D. Figure 11 B is a diagram illustrating the variation between groups of six groups, as shown in Figures 11A and 11C. Figure 11C is a diagram that is referred to as a dendrogram to illustrate each pair of group centroids using the distance matrix, as found in Figure 8. Figure 11 D is a diagram of a dendrogram generated during a repetition of the dendrogram generated, as shown in Figures 8 and 11 B. Figure 12 is an example of the HELTR table of the assessment of the six groups, as shown in Figure 11D.

DETAILED DESCRIPTION OF THE INVENTION In Figure 1 a block diagram is shown in high-level view of the valuation of assets using the indistinct grouping process. As shown, and discussed in more detail later, the process includes acquiring a number of 2A-2Z loan portfolios. These loan portfolios are then restructured using a portfolio restructuring process 3. The restructured portfolios are then fed into a merged loan information process 4, which takes information from a loan information database 5 and . .m.,. M \ mi1f¡ &? T. j? UttMtt.m. ...... ..? *** H? Íjati,? M. .? .. mmMtlm? m merges that loan information with the restructured portfolios. The portfolios restructured with the loan information added 5 afterwards come out as portfolio restructures 6A- 6H. These restructured portfolios 6A-6H are then fed into a (expand) HELTR process 7 of the present invention. The HELTR process 7 provides a method for forming an asset valuation using the indistinct grouping of the present invention. The valuation of the assets using the indistinct grouping method of the present invention takes the aforementioned loan data, then results in the expected cash flow and risk valuation of each of the asset loan portfolios. illustrated in Figure 2, a computer system 21 generally comprises a processor 22 and memory 31 (e.g., RAM, ROM, hard disk, CD-ROM, etc.) with an operating system 32. Processor 22 accepts code and data of the memory 31 of the local interface 23, for example, a port concentrator (s). The user address can be signaled through the use of input devices, for example but not limited to, a mouse 24 and a keyboard 25. The resulting input and output actions are displayed on a display terminal 26. An evaluation of Active using the indistinct group system 50 can access other computers and resources in a network, using a modem or network card 27. An asset valuation is also shown in Figure 2 using the indistinct grouping system 50 which includes: a data acquisition process 60, a variable selection process 80, a hierarchical segmentation process 100, an indistinct grouping process 120, and a subscription revision process 180, which they are in the memory area 31. The databases 33 are also shown to reside the memory area 31. These components are described here in further details with reference to Figures 2-12. The memory area 31 may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, apparatus, device or means of propagation. More specific examples (a non-exhaustive list) of the memory area 31 include any one or more of the following: an electrical (electronic) connection having one or more cables, a portable (magnetic) diskette, a random access memory (RAM) (magnetic), a read-only memory (ROM) (magnetic), an erasable programmable read-only memory (EPROM or Flash memory) (magnetic), an optical fiber (optical), and a portable compact disk of read-only memory (CDROM) (optical). An example of the system and flow method for asset valuation using the indistinct grouping system 50 of this invention is illustrated in Figure 3. The following description of the system and method for valuation of assets using the indistinct grouping system 50 uses the example of loan portfolios. Assets can be, but are not limited ^^ gj ^ j g ^^ j ^ mé ü a, loans, insurance policies, bonds, stocks, property leases, and other properties. The system and method for asset valuation using the indistinct grouping system 50 includes the data acquisition process 60 which comprises the step of acquiring data. For the example used in the invention: loan portfolios, loan history information, including settled data of loan payment history, settled credit analysis data, settled data of car loans and mortgage loans, and settled data of specific industry. The data acquisition process 60 employs a divide and beat strategy to handle massive amounts of asset data. Asset data is captured in the variable selection process 80 that identifies the critical loan variables for credit review or those variables with the greatest discriminatory power and separates the various loan groups. The loan portfolio data collected in the acquisition process 60 and the critical variables identified in the variable selection process 80, both are captured within the hierarchical segmentation process 100. The hierarchical segmentation process 100, divides the entire portfolio of assets (ie, for example, loans) in a number of deposits based on a predetermined critical variable through that solvency review of the portfolios. After the hierarchical segmentation process 100 is carried out, the divided assets (i.e., loans) are further classified through the indistinct grouping process 120. The indistinct grouping process 120 classifies each of the deposits divided into one. predetermined number of groups based on a natural structure of the asset data. After this classification through the indistinct grouping process 120 is carried out, the classification of assets (ie, loans) is also processed through the financing review process 180 that allocates cash flow and risk scores. projected for the assets for each of the groups. The projected cash flow and risk scores for the assets of the groups give results to be used in the solvency analysis of the portfolios. The data acquisition process 60, the variable selection process 80, the hierarchical segmentation process 100, the indistinct grouping process 120 and the financing review process 180 are defined here with additional details according to Figures 4-12 . In Figure 4 there is illustrated a flow diagram of the example of the valuation of assets using the indistinct grouping system 50. First, the valuation of assets using the indistinct grouping system 50 executes the data acquisition process 60 in step 51 The data acquisition process 60 is defined here with additional details according to Figure 5 The valuation of assets using the indistinct grouping system 50 then executes the variable selection process 80 in f JBU, .A. * A ...- rtim-u. -Ato. *. * A "step 52. The variable selection process 80 uses the data of assets captured in the data acquisition process 60. The variable selection process 80 is defined here with additional details with reference to Figure 6. The valuation of assets using the indistinct grouping system 50 then executes the hierarchical segmentation process 100. The hierarchical segmentation process 100 divides the asset portfolio, i.e., loans, into a predefined user number of groups based on the critical variables of the identified user The hierarchical segmentation process 120 is defined herein with additional details in accordance with Figure 7. The valuation of assets using the indistinct grouping system 50 then executes the indistinct grouping process in step 54. The indistinct grouping process 120 further classifies each of the groups in the hierarchical segmentation process 100 into a predetermined number of user groups based on the natural structure of the asset data. The indistinct grouping process 120 is defined here with additional details in accordance with Figure 8. The valuation of assets using the indistinct grouping system 50 executes the financing review process 180 in step 55. The financing review process 180 allocates the projected cash flow and risk scores to each of the groups identified by the indistinct grouping process 120. The financing review process 180 is defined here L :. .. . ^ at¡taJí &. ^ ?. El.i? with further details with reference to Figure 10. In step 59, the asset valuation using the indistinct pool system 50 then exits. Figure 5 illustrates a block diagram illustrating types of asset database examples used in the construction of the appendix of the relevant asset variables to the asset portfolio. The data acquisition process 60 includes the passage of data related to the acquisition of assets. This step usually involves attaching relevant data to the assets in the asset portfolio. This invention will illustrate these concepts using an example of a loan discussed through this invention. Using this example, the loans will have relevant cross-referenced data and merged into the asset information. Preferably, the loan asset data in the loan database 60 is cross-referenced and merged with multiple databases. For example, as shown in Figure 5, the example loan database may include records from a variety of different universal files or databases including, for example, but not limited to, settlement of payment history data of loan 35, settlements of data of analysis of public credit 36, data of settlements of data of hypothecating loans and loans for automobiles 38, and establishment of specific data of industry 39. Preferably, the registers of assets of loan in the database of loan assets 60 f £ A? * $ - tt < »Rí -? - ** -" k ?? * - émm ^ .. ...- -Jma * - - • - "" - «J ^ *» - 1 - ^^ -..- «« ..... nt * -.,. they are fused with the data previously referenced which will be useful in identifying critical variables that are used during the hierarchical segmentation process 100 and the execution of the indistinct grouping process 120. Before deducting useful information outside the composition of the database , a data purification is carried out in the collected data. Therefore, data debugging includes but is not limited to, detecting errors, filling or deleting lost values, deriving variables captured from raw data, etc. In Figure 6 a table is illustrated of an example of an implementation of the variable selection process 80. In the variable selection process 80, the user identifies those variables that are estimated critical. In the present example, the variable selection process 80 has identified 11 variables that will be used by the indistinct grouping process 120. As seen, associated with each of the critical estimated variables there is an associated category and / or value scale for the variable, as well as a coding scheme to represent the variable. Figure 7 illustrates an example of a hierarchical segmentation model created during the hierarchical segmentation process 100 with respect to the example of loan portfolios. An example of the hierarchical segmentation model applied through the hierarchical segmentation process 120 is CARJ. CART is a well-known statistical algorithm of regression trees and is used for hierarchical segmentation. The idea behind the regression trees is to segment the loan portfolios into a predetermined number of categories such that each category is homogeneous with respect to the user's default critical variables. Figure 7 shows the result of the application of the hierarchical segmentation process of the CART model. The resulting regression tree has the example of segmented loan portfolios using three critical variables. These three critical variables for this example include loan security, the type of loan and the last loan payment. The three regression partitions resulting from the loan portfolio using these three critical variables within six groups. These partitions can be represented as a tree structure using the CART model. Once the portfolio is segmented by using critical variables, the indistinct grouping process 120 can carry out a fine granular partition of each of these predetermined groups. In Figure 8 a flow chart of an example of an implementation of the indistinct array process 120 of the present invention is illustrated. The implementation example shown in Figure 8 uses the example of the segmented loan portfolio generated by the hierarchical segmentation process 100. The indistinct grouping process 120 is first initiated in step 121. Then, in step 122, the process of indistinct grouping calculates the means of indistinct grouping by means of the -HJ i -t -.... ^ Rm ^ M ^^ Uá .-. ^. I? .-. ^ ^ .......-... át ^^, ^, ^. ^ i ^ m ^ m ^ r ^^ ü? iM execution of a calculation FCM process 140. The computation FCM process 140 is defined here with further details with reference to Figure 9. In step 123, the indistincting process 120 calculates the variation between groups and within groups through the argument box. This calculation of the variation between groups and within groups by means of the argument box is a diagnostic check in the final result. The diagnostic check is carried out by examining the corresponding argument boxes between groups and within groups, respectively. The argument boxes of variation between groups and within groups are defined herein with additional details with reference to Figures 11 (A &B). In step 124, the indistinct grouping process 120 determines whether the variation between groups and within groups is sufficiently compact or there is only one more group. The determination of whether the grouping is sufficiently compact is carried out by determining whether the variation within groups is minimized while the variation between groups is maximized by the calculation FMC process executed in step 122. If it is determined in step 124 that the grouping is compact enough or that only one is left over, the indistinct grouping process 120 goes out in step 139. If it is determined that the grouping is not compact enough in step 124, the indistinct grouping process 120 then take the first pair in the next group of centroids of . -, ... ^ A ^ i ..OLÍ-,. «** fa» -t tU? M. t.maMr? mJ * kÍ ... group in step 125. The indistinct grouping process 120 calculates the distance between each pair of group centroids and stores this distance in a distance matrix in step 126. In step 131, the indistinct grouping process 120 determines whether or not there are more pairs of centroids of groups to be examined. If there are no more pairs of centroids of groups to be examined, the indistinct grouping process 120 repeats steps 125-131. If there are no more pairs of group centroids to be processed, the indistinct grouping process 120 then generates a dendrogram for each pair of group centroids using the distance matrix in step 132. The indistinct grouping process 120 in step 133 inspects the dendrogram for possible mergers of group centroids. The example of a dendrogram is defined herein with additional details and with reference to Figures 11A to 11D. In step 134, the indistinct grouping process 120 determines whether any of the group centroids can be merged. If there is a possible fusion of a pair of group centroids, the indistinct grouping process 120 returns to repeating steps 122 to 124. If the indistinct grouping process 120 determines that there is no possibility of fusion of group centroids, the process of indistinct grouping 120 then exits in step 139 This process is defined here with additional details and with reference to Figures 11 A to 11B.

In Figure 9 the calculating process FCM 140 is illustrated. First, in step 141, the number of heavy groups or exponents are entered into the calculation process FCM 140. Then, the calculation FMC process 140 obtains the first group in step 142. In step 143, the first (next) data point is obtained. The calculation process FMC 140 then randomizes the degree of grouping of each point of each group in step 144. The degree of grouping μík is defined by Intuitively, μ, k, the degree of grouping of the data point Xk in the centroid of the group V ,, will become greater as Xk and closer to V ,. At the same time, μ, k, will become smaller as Xk and farther from V, (other groups). In step 145, the calculation process FMC then determines whether all data points in the current group have been scrambled in step 145. If it is determined in step 145 that all data points have not been scrambled, the process FMC calculation 140 then repeats steps 143 to 145. If the calculation FMC process 140 determines that all data points for the current group have been scrambled, the calculation FMC process 140 then determines whether all points data have been randomized for all variable groups in step 146. If the calculation FMC process 140 determines that not all groups have had their data points scrambled, then the calculation FMC process 140 repeats steps 142 to 146. If the calculation process FMC determines that all the data points of all the groups have been alloyed, the calculation FMC process 140 then calculates the centroid for all the data points in step 147. The center group V, is defined by Instinctively, V ,, is the group centroid, is the heavy sum of the coordinates of Xk, where k is the number of data points Beginning with the desired number of groups c and an initial speculation for each group center V ,, / = 1,2,, c, the FMC calculation process 140 will converge to a solution for V ,, which represents both a local minimum and a peak point of the cost function The quality of the solution of the calculation FMC process 140, like most nonlinear organization problems, depends heavily on the selection of initial values-the number c and the initial group centroid V, then, in step 148, the calculation process FMC 140 LAA -. * I * «| illi1il1ta - »- - - .- ^^^« ^ »« A - ^ - »« »a ^ * - ^ = fa ^«. «^ s ^^" ^ - ^ - - • s- < ttfeáh '* - s calculates the objective function The objective function is defined by: - where n is the number of data points; c is the number of groups, Xk is the kavo data point; Vi is the group centroid avo; μlk is the degree of clustering of the data in the group; m is a constant greater than 1 (typically m = 2). Note that μlk is a real number and defined in [0,1]. μlk = 1 means that the data is definitely in the kav group, while μlk = 0 means that the data is definitely not in the group. If μlk = 0.5, then it means that the data is partially in the kavo group to the 0.5 degree. Instinctively, the cost function should be minimized if each data point belongs exactly to a specific group and there is no partial degree of grouping in any other group. That is, there is no ambiguity in assigning each data point to the group to which it belongs. In step 149, the calculation process FMC 140 determines whether the value of the objective function of calculation is convergent. If the objective function of calculation is non-convergent, the calculation process FMC 140 proceeds to steps 151 to 155. If the calculation FMC process 1400 determines that the value of the objective function is convergent in step 149, the FMC process of calculation then goes out in step 159.

In step 151, the calculation FMC process 140 obtains the first group. In step 152, the first data point is obtained. The calculation process FMC then updates the degree of grouping of each data point in each group in step 153. In step 154, the calculation process FMC 140 determines whether each data point in the current group has been updated. If all the data points in the current group have not been updated, the calculation FMC process 140 then repeats steps 152 to 154. If all the data points of the current group have been updated, the calculation FMC process 140 it then determines whether each data point of all the groups has been updated in step 155. If the calculation FMC process 140 determines that not all the data points have been updated for all the groups in step 155, the FMC process of calculation 140 then repeats steps 151 through 155. If the calculation process FMC 140 determines that all the data points of all the groups have been updated in step 155, the calculation FMC process then repeats the steps 147 -149, as defined above. Figure 10 illustrates a financing review process. The financing review process is carried out after the entire loan portfolio is segmented through the indistinct grouping process 120. During the 180 review process, each group is reviewed and assigned a composite score called HELJR . HELJR means: H = high cash flow; E = expected cash flow; L = low cash flow; T = timing of cash flow in months; and R = risk assessment of the lender. In essence, the HELTR score captures both the expected cash and the scale, the timing of the cash flow and the risk associated with each group. First, the financing review process 180 is initiated in step 181. In step 182, the first loan segment is obtained and makes the second loan segment. In step 183, the first group in the current loan segment is obtained. In step 184, the financing review process 180 calculates the cash flow score and the timing of each cash flow for the current group in the current loan segment. In step 185, the financing review process 180 calculates the risk assessment in the purchase of all groups in the current loan segment. In step 186, the review process determines whether the valuation of all groups in the current loan segment has been carried out. If there are no groups in the current loan segment, the funding review process 180 repeats steps 183 through 186. In step 187, the funding review process 180 determines whether all groups have been reviewed in all of the loan segments. If the funding review process 180 determines that not all groups in all loan segments have been reviewed, the 180 review process of funding again repeats steps 182 through 187 if the funding review process 180 determines that all the groups in all the loan segments have been reviewed, the financing review process 180 then goes out in step 189. 5 Figures 11A and 11B illustrate the variation between -groups and within groups of an example of six groups generated in step 123 (Figure 8). As shown in Figure 11 B, the average distance of all data points to the centroid is 1.0, while the average distance of the group data points is 0.6. 10 as shown in Figure 11 A. This indicates that the grouping is fairly compact. Therefore, the variation between groups is minimized, while the variation within groups is maximized by the calculation process FMC 140. Figures 11C and 11D illustrate dendrograms of 20 15 groups and 6 groups, respectively. It is observed that each dendrogram begins with a data point forming a separate group, and where the data points or groups close to each other are successfully merged. This usually requires a couple of iterations of the indistinct grouping process 120 to obtain the optimum number of groups. As shown in Figure 11C, of the 20 centroids, 12 of them must be merged. Therefore, in the previous example, the indistinct grouping process 120 is then re-executed with 6 groups with the result being shown in the Figure 11 D. 25 Figure 12 illustrates an example of a HELTR chart for the 6 groups shown in Figure 11D. As shown, the HELTR table includes data for each centroid identified in the example in Figure 7 and for each group within each identified centroid. The data used in the HELTR chart include whether the loans are insured or not, revolving credit, required notice, including the last payment, maturity of the loan, or whether the loans are secured or not. The collective score is shown, tax position, principal balance not paid in millions, total percentage of principal not paid, and cash flow analysis. The method and system for the valuation of assets using the indistinct grouping system 50 comprises an ordered list of executable instructions for implementing logic functions. The ordered list may be contained in any computer-readable medium for use by or in connection with a system for executing instructions, apparatus or device, such as a computer-based system, system containing a processor, or other system that may take up instructions of the instruction execution system, device, or device, and execute the instructions. In the context of this document, a "computer-readable medium" may be any means that may contain, store, communicate, propagate, or transport the program to be used by or in connection with the instruction, device or device execution system. The computer-readable medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, apparatus, device or means of propagation. More specific examples (a non-exhaustive list) of the computer-readable medium may include the following: an electrical (electronic) connection having one or more cables, a laptop (magnetic) diskette, a random access memory (RAM) (magnetic) ), a read-only memory (ROM) (magnetic), an erasable programmable read-only memory (EPROM or Flash memory) (magnetic), an optical fiber (optical), and a portable read-only compact disk (CDROM) ) (optical). Note that the computer-readable medium can also be paper or other suitable medium on which the program is printed, since the program can be captured electronically by, for example, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in an appropriate way if necessary, and then stored in the computer's memory. The above description has been presented for purposes of illustration and description. It does not attempt to be exhaustive or limit the invention to the precise forms described. Obvious modifications and variations are possible in light of the previous teachings. The flow diagrams of the invention show the architecture, the collection of functionality and interpretation system. In this way, each block represents a module, segment, or portion of ? Ü code, which comprises, one or more executable instructions to implement the specific logical function (s). It should also be noted that in some alternative implementations, the functions observed in the blocks may occur out of service in the figures, or for example, in fact be executed substantially or in reverse order, depending on the functionality involved. The system and methods discussed were selected and described to provide the best illustration of the principles of the invention and their practical application to enable one skilled in the art to use the invention in various modalities and with various modifications as are appropriate to the invention. private use contemplated. All modifications and variations within the scope of the invention as determined by the appended claims when interpreted to the extent to which they are clearly and legally entitled. ^ fáW |! l == í

Claims (21)

1. CLAIMS 1. A method for providing an asset valuation using the indistinct grouping system comprising the steps of: acquiring a plurality of assets and attaching a plurality of variables to said plurality of assets; examine said plurality of assets and said plurality of variables with a model; use indistinct grouping to calculate the value of said plurality of assets and said variables; and calculate a return on said plurality of assets. The method according to claim 1, wherein the step of using indistinct grouping further comprises the step of: randomizing a degree of grouping of each of the plurality of data points in each of the plurality of groups; and calculating a centroid for said plurality of data points and said plurality of groups. The method according to claim 2, wherein said step of using indistinct grouping further comprises the step of: updating a degree of grouping for each of said plurality of data points in each of said plurality of groups 4. The method according to claim 3, wherein the step of using indistinct grouping further comprises the step of: calculating a value with an objective function; and determine if said objective function value is convergent. The method according to claim 2, wherein the step of using indistinct grouping further comprises the steps of: calculating variants between groups; and calculate variants within groups. The method according to claim 5, wherein the step of using indistinct grouping further comprises the step of: compacting said variants between groups and said variants within groups. The method according to claim 6, wherein the step of compacting said variants between groups and said variants within groups further comprises the steps of: calculating the distance between each of said variants between groups and said variants within groups; storing the distance between each pair of said variants between groups and said variants within groups in a distance matrix; generate a dendrogram for the distance between each pair of said variants between groups and said variants within groups; and evaluate the dendrogram for possible mergers. 8. A system that provides asset valuation using .? JUAA A *. »^ - ^^^. ^^^ mmj ^ ^. ^ ^ Indistinct grouping comprising: means to acquire a plurality of assets; means for attaching a plurality of variables to said plurality of assets; means for examining said plurality of assets and said plurality of variables with a model; means for using indistinct grouping to calculate the value of said plurality of assets and said variables; and means to calculate a return on said plurality of assets. The system according to claim 8, wherein said indistinctly grouping means further comprises: means for randomizing a degree of grouping of each of the plurality of data points in each of the plurality of groups; and means for computing a centroid of said plurality of data points and said plurality of groups. The system according to claim 9, wherein said indistinctly grouping means further comprises: means for updating a degree of grouping each of the plurality of said data points in each of said plurality of groups. The system according to claim 10, wherein said indistinctly grouping means further comprises: means for calculating a value with an objective function; Y means to determine if said value is convergent. The system according to claim 8, wherein said indistinctly grouping means further comprises: means for calculating variants between groups; and means to calculate variants within groups. The system according to claim 12, wherein said indistinctly grouping means further comprises: means for compacting said variants between groups said variants within groups. The system according to claim 13, wherein said indistinctly grouping means further comprises: means for calculating the distance between each pair of said variants between groups and said variants within groups; means for storing a distance between each pair of said variants between groups and said variants within groups in a distance matrix; means for generating a dendrogram for said distance between each of said variants between groups and said variants within groups; and means to evaluate the dendrogram for possible mergers. 15. A system for providing profitability stability analysis for different types of goods or services comprising: logic of data acquisition that acquires a plurality of assets and attaches a plurality of variables of said plurality of assets; segmentation logic that examines said plurality of assets and said plurality of variables with a model; indistinct grouping logic that calculates a value of said plurality of assets and said variables; and logic of profitability analysis that calculates a profitability of said plurality of assets. 16. The system according to claim 15, wherein the indistinct grouping logic further comprises: logic that randomizes a degree of grouping of each of the plurality of data points in each of the plurality of groups; and logic that calculates a centroid for said plurality of data points and said plurality of groups. 17. The system according to claim 16, further comprising: logic updating a degree of grouping each of said plurality of data points in each of said plurality of groups. 18. The system according to claim 17, further comprising: logic for calculating a value with an objective function and determining whether the value is convergent. 19. The system according to claim 17, further comprising: logic for calculating variants between groups of said plurality of data points in each of said plurality of groups; and logic for calculating variants within groups of said plurality of data points in each of said plurality of 5 groups. 20. The system according to claim 19, further comprising: logic compacting said variants between groups and said variants within groups. 21. The system according to claim 16, further comprising: logic for computing a distance between each pair of said • variants between groups and said variants between groups; storage logic of a distance between each pair of said 15 variants between groups and said variants between groups in a distance matrix; logic for generating a dendrogram for said distance between each pair of said variants between groups and said variants within groups; Y • 20 evaluation logic of the dendrogram for possible mergers. 25 jal j- | r "Maii '-'- - m ^ í¡m ^ .. ^ mtt.mmm4