JP2003076860A - Reinsurance evaluation method and computer system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate the effect of reinsurance in a more appropriate manner. SOLUTION: An allotment scheme change provisional calculation processing part 52 changes from a reinsurance allotment scheme applied to actual insurance contract corresponding to a claim for an actually occurred damage into a designated reinsurance allotment scheme, and calculates, for each claim, according to the designated reinsurance allotment scheme, a recovery insurance money due to the reinsurance. A random provisional calculation processing part 53 associates the claim for the actually occurred damage with an arbitrary insurance contract, calculates, for each claim, a recovery insurance money, its expected value or the like due to the reinsurance according to the designated reinsurance allotment scheme applied to the insurance contract associated with each claim. An insurance money simulation processing part 54 generates the claim for each insurance contract according to the information on occurrence frequency and expected amount of damage generated on the basis of the information on the actually occurred damage, and calculates, for each claim, the recovery insurance money by reinsurance, its expected value or the like according to the designated reinsurance allotment scheme.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a reinsurance evaluation technique in an insurance company.

[0002]

2. Description of the Related Art Reinsurance is to transfer some or all of the insurance liability undertaken by an insurance company to another insurance company. As a result, it is possible to stabilize the management of the insurance company by complementing one's own underwriting capacity and at the same time diversifying and leveling the risks underwritten. However, the burden of reinsurance charges for reinsurance arrangements is a large cost,
It is necessary to use the reinsurance system in an appropriate manner by balancing the amount of reinsurance that is expected to be collected by reinsurance and the reinsurance premium required for the reinsurance arrangement. Therefore, for each prescribed form of insurance contract, the insurance company itself underwrites the insurance liability (called holding), which is retained within the company without transferring to other companies, and the amount of reinsurance (called cedre). A reinsurance allocation scheme for the reinsurance is determined in advance and, for example, the amount of reinsurance collected within a predetermined period (hereinafter referred to as “recovered insurance money.” In the present application, the reinsurance fee that can be collected by ceding may be included. ) Was evaluated, and the reinsurance allocation scheme for the next period was changed.

[0003]

However, the conventional evaluation methods do not perform rigorous calculations, and the evaluation method is not sufficient, so that the effect of reinsurance cannot be evaluated more appropriately. It was

[0004] Therefore, an object of the present invention is to provide a technique for making it possible to more appropriately evaluate the effect of reinsurance.

[0005]

A reinsurance evaluation method according to a first aspect of the present invention associates generated claim data for a claim that actually occurred with arbitrary insurance contract data and stores a correspondence relationship. The reassociation step stored in the device and the reinsurance allocation scheme to be applied to the insurance contract data associated in each associating claim data in the associating step are determined, and each reinsurance allocation scheme is determined according to the determined reinsurance allocation scheme. Calculating a reinsurance recovery amount for the claim and storing the reinsurance recovery amount in a storage device.

By associating the claim data with arbitrary insurance contract data in this way, the claim data can be associated with insurance contract data different from the actual insurance contract data. In that case, damage such as an accident that actually occurred will occur to other insurance contracts, and by calculating the reinsurance claim recovery amount at that time, data for evaluating the effect of reinsurance can be obtained. Like

Further, in the associating step described above, the generated claim data may be associated with the insurance contract data based on a random number. As a result, the damage due to an accident or the like can be caused to the insurance contract with an equal probability, for example.

In the first aspect of the present invention, the method further includes a step of categorizing the insurance contract data to be processed and the generated claim data based on the category information designated as the trial calculation condition, and the above-mentioned association is performed. The step may be a step of associating, for each category, the generated claim data included in the category with arbitrary insurance contract data included in the category and storing the correspondence relationship in the storage device. By categorizing, it is possible to reduce the combination of insurance contracts and claims that cannot normally occur in the rearrangement of the correspondence relationship, and it becomes possible to carry out a more appropriate evaluation.

When the category information described above is specified by at least one of an insurance target property, a location, an insurance amount, job work, and in the case of joint insurance, the contents of the joint insurance. There is also.

Further, in the first aspect of the present invention, the associating step and the reinsurance claim recovery calculation step described above are repeatedly performed, and the expected value of the total reinsurance claim recovery value for all claims is repeated in each iteration. May be calculated and stored in a storage device. This expectation allows us to see the average effect of reinsurance on the reinsurance allocation scheme.

Further, in the first aspect of the present invention, the reinsurance recovery amount for each claim calculated in the above-mentioned reinsurance amount recovery calculation step is an insurance contract for determining a reinsurance allocation scheme. In some cases, the method may further include the step of totaling by pattern. It is an index for evaluating the effect of reinsurance.

Further, in the first aspect of the present invention, the method further includes the step of totaling the paid insurance claims and the unpaid insurance claims recorded for one damage to generate claim data, and storing the claim data in a storage device. Various configurations are also possible. This is to deal with the case where the paid insurance amount and the unpaid insurance amount are separately recorded for one damage.

Further, in the first aspect of the present invention, the reinsurance premium is calculated and stored based on the reinsurance allocation scheme applied to the insurance contract data associated with each claim data in the associating step. The configuration may further include the step of storing in the device. This is because it is necessary to calculate the cost for reinsurance arrangement.

Further, in the above-mentioned reinsurance claim recovery amount calculation step, the reinsurance allocation scheme applied to the insurance contract data associated in the association step with each incurred claim data is set to an arbitrary reinsurance scheme. The configuration may be such that it is determined using the input information regarding the insurance allocation scheme. The reinsurance allocation scheme to be applied can be applied not only to the reinsurance allocation scheme actually applied but also to the reinsurance allocation scheme to be adopted or other reinsurance allocation schemes.

Further, in the first aspect of the present invention, the reinsurance allocation scheme applied to the actual insurance contract data corresponding to the claim data about the actually caused damage is changed to the designated reinsurance allocation scheme. The reinsurance claim recovery amount may be calculated for each claim according to the designated reinsurance distribution scheme and stored in a storage device. Calculate the reinsurance recovery amount according to the specified reinsurance allocation scheme without changing the policy and claim data pairs from the actual ones. Then, it becomes possible to compare with the case where the pair recombination is performed.

Further, in the first aspect of the present invention, the expected complaint data is generated according to the information of the occurrence frequency and the expected damage amount generated based on the information about the actually occurring complaint, and the expected complaint data is used as the insurance contract data. The method may further include the step of calculating the reinsurance claim recovery amount for each expected claim in accordance with the specified reinsurance allocation scheme and storing it in a storage device. Virtually incur damage due to an accident, etc., and calculate the reinsurance claim recovery amount by the specified reinsurance allocation scheme.
As a result, the reinsurance claim recovery amount can be calculated in consideration of a possible accident that has not actually occurred.

The reinsurance evaluation method according to the second aspect of the present invention corresponds to a step of storing input information relating to an arbitrary reinsurance allocation scheme in a storage device, and generated claim data regarding a claim actually generated. Determining the reinsurance allocation scheme to be applied to the actual insurance contract data using the input information, calculating the reinsurance recovery amount for each claim according to the determined reinsurance allocation scheme, and storing it in the storage device. including.

The reinsurance evaluation method according to the third aspect of the present invention is based on the information about the claim actually occurred.
Generates frequency of occurrence and expected damage amount data, and stores the frequency / damage amount in the storage device, and generates expected complaints based on the occurrence frequency and expected damage amount data, and generates expected complaint data for each expected complaint Then, the expected claim data generation step of storing each expected claim data in the storage device, the association step of associating the expected claim data with the insurance contract data and storing the correspondence relationship in the storage device, and any reinsurance Using the input information, the allocation scheme storing step of storing the input information about the allocation scheme in the storage device and the reinsurance allocation scheme applied to the insurance contract data associated with each expected claim data in the association step And decide
Calculating a reinsurance payback amount for each expected claim according to the determined reinsurance allocation scheme and storing the reinsurance payback amount in storage.

In the frequency / damage amount generation step described above, the frequency of occurrence and the expected damage amount data are generated by setting the frequency distribution and the expected damage amount distribution on the basis of the information relating to the actually occurring complaint. Such a configuration may be adopted. Further, in the above-described expected complaint data generating step, the expected complaint may be generated based on a random number using the occurrence frequency and the expected damage amount data.

The reinsurance evaluation method according to the fourth aspect of the present invention comprises a reinsurance allocation scheme storing step of storing input information regarding an arbitrary reinsurance allocation scheme in a storage device,
According to the reinsurance allocation scheme based on the input information, which relates the generated claim data for the actually generated claims to any insurance contract data and is applied to the insurance contract data related to each generated claim data. Expected claims according to the first reinsurance claim recovery calculation step of calculating the reinsurance claim recovery amount for each claim and storing it in the storage device, and the occurrence frequency and the expected damage amount information generated based on the information on the claim actually occurred.・ Each expected claim according to the reinsurance allocation scheme based on the input information, which generates the data, associates the expected claim data with the insurance contract data, and applies the insurance contract data associated with each expected claim data. Second reinsurance claim recovery calculation step for calculating the reinsurance claim recovery amount and storing it in the storage device Including the.

According to this, the combination of the actual accident and the insurance contract data is rearranged to generate the reinsurance amount calculated according to the designated reinsurance allocation scheme and the damage caused by the accident. , Will be able to compare with the reinsurance payout calculated according to the specified reinsurance allocation scheme.

Further, in the above-mentioned first and second reinsurance claim recovery amount calculation steps, the reinsurance claim recovery amount calculated for each claim may be summed up.

The reinsurance valuation method according to the fifth aspect of the present invention is the reinsurance allocation specified from the reinsurance allocation scheme applied to the actual insurance contract data corresponding to the claim data regarding the actually caused damage. Change to the scheme, calculate the reinsurance claim recovery amount by reinsurance for each claim according to the specified reinsurance allocation scheme, and store it in the storage device, first recovery claim calculation step, and the claim that actually occurred claim -Associating the data with any insurance contract data, calculating the reinsurance recovery amount for each claim according to the specified reinsurance allocation scheme applied to the insurance contract data associated with each incurred claim data, The second collection insurance claim calculation step stored in the storage device, and the occurrence frequency generated based on the information regarding the actually generated complaint And generate expected claim data according to expected loss information, associate expected claim data with insurance contract data, and specify reinsurance applied to insurance contract data associated with each expected claim data Calculating a reinsurance recovery amount for each expected claim according to the allocation scheme and storing it in a storage device.

In the first recovery insurance money calculation step, the recovery insurance money for the designated reinsurance allocation scheme for one that actually occurs, which is one form of damage due to an accident that may occur, can be obtained. In the second recovery insurance money calculation step, it becomes possible to obtain the recovery insurance money and the expected value thereof for the designated reinsurance allocation scheme in the case where the mode of occurrence of damage due to an actual accident or the like is changed. In the third recovery insurance money calculation step, it becomes possible to obtain the recovery insurance money and the expected value thereof for the designated reinsurance allocation scheme in consideration of the damage caused by an accident that has not yet occurred. By comparing these, it becomes possible to properly evaluate the effect of reinsurance.

In some cases, the first to third recovery insurance money calculation steps described above may include a step of summing the reinsurance recovery amount calculated for each claim. Further, the second and third collected insurance money calculation steps may include a step of calculating an expected value of the total collected insurance money.

It is also possible to create a program for causing a computer to execute the reinsurance evaluation method according to the present invention. The program is, for example, a flexible disk, a CD-ROM, a magneto-optical disk, a semiconductor memory, a hard disk. Etc. are stored in a storage medium or a storage device. It may also be distributed via a network. The data being processed is temporarily stored in the memory of the computer.

[0027]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A system outline according to an embodiment of the present invention will be described with reference to FIG. In the calculation center 3,
Provision of normal insurance contracts and claims (herein comprehensively means accidents, details of accidents, damages, damages, amounts covered by insurance, etc.) and a normal system 31 for processing reinsurance Information regarding reinsurance is stored in the reinsurance DB 33, information regarding insurance contracts is stored in the insurance contract DB 34, and claim data regarding damage caused by accidents and the like is stored in the claim DB 35. It is assumed that the reinsurance DB 33 also stores information on the actually applied reinsurance distribution scheme, a pattern code setting table described later, and the like. The calculation center 3 extracts the trial calculation data used by the trial calculation server 5 from the insurance contract DB 34, the reinsurance DB 33, and the complaint DB 35, and transmits the trial calculation data to the trial calculation server 5, and the trial calculation data creation / transmission server 37 and the trial calculation data. A trial calculation data storage unit 39 for temporarily storing the trial calculation data generated by the data creation / transmission server 37 is also installed. For example, creating trial calculation data
The transmission server 37 may periodically extract the trial calculation data from the insurance contract DB 34, the reinsurance DB 33, and the complaint DB 35 and store the trial calculation data in the trial calculation data storage unit 39. Sometimes insurance contract DB
34, the reinsurance DB 33 and the complaint DB 35 may be extracted for trial calculation data and transmitted to the trial calculation server 5.

The calculation center 3 and the trial calculation server 5 are connected to the trial calculation server 5 via a network 1 which is, for example, a public line network (which may be a LAN (Local Area Network) or the Internet). The trial calculation server 5 includes a trial calculation processing unit 51 including an interface with the user terminal 7.
And the trial calculation data creation / transmission server 37 of the calculation center 3
And a download processing unit 55 for downloading the trial calculation data from. The download processing unit 55 may be configured to periodically request the trial calculation data from the trial calculation data creation / transmission server 37.
When performing the process in 1, the trial calculation data creation / transmission server 37 may be requested for the trial calculation data. The trial calculation processing unit 51 includes a distribution scheme change trial calculation processing unit 52, a random trial calculation processing unit 53, and an insurance money simulation processing unit 54. The processing contents of these processing units will be described in detail later. In addition, the trial calculation server 5 includes a trial calculation data storage unit 5 that stores trial calculation data.
6, the trial calculation condition data storage unit 57 that stores the trial calculation condition data transmitted from the user terminal 7, and the trial calculation result storage unit 58 that stores the trial calculation result by the trial calculation processing unit 51.

One or more user terminals 7 are connected to the trial calculation server 5 via the network 2. User terminal 7
May be installed at any place (for example, a head office, a branch office, a group company, etc.). Further, it may be installed in a plurality of places. The network 2 is, for example, a LAN
Is. However, the present invention is not limited to this, and may be a public line network or the Internet. User terminal 7
Is a personal computer, for example, which executes a dedicated program as an interface with the trial calculation server 5, or exchanges information with the trial calculation server 5 which functions as a Web server using, for example, a Web (Web) browser. In some cases.

Next, the outline of the processing flow of the trial calculation server 5 will be described with reference to FIG. As a premise of FIG. 2, the download processing unit 55 of the trial calculation server 5 receives the trial calculation data from the trial calculation data creation / transmission server 37 of the calculation center 3, and the trial calculation data storage unit 56 managed by the trial calculation server 5.
Stored in. For example, the data for trial calculation is
Securities number, insurance type, insurance type, property type, job work code, location, contractor name, contract start date, contract end date, insurance amount (allocation reference amount), and co-insurance classification (single / co-insurance type and content etc. ), Etc. and accident number, accident date, accident date, insurance type, insurance type, property name, job code, location, contractor, security number, insurance claim, payment /
Includes complaint data including unpaid information. In addition,
It may also include information such as a scheme content table including the content of the reinsurance allocation scheme actually applied and a pattern code setting table that associates the number (code) of the reinsurance allocation scheme from the insurance contract content.

In FIG. 2, the distribution scheme change trial calculation processing unit 52 of the trial calculation processing unit 51 executes the distribution scheme change trial calculation processing according to the trial calculation condition data and the trial calculation instruction from the user terminal 7 (step S1). Also, the trial calculation processing unit 5
The random trial calculation processing unit 53 of No. 1 executes the random trial calculation processing in accordance with trial calculation condition data and trial calculation instruction from the user terminal 7 (step S3). Then, the insurance money simulation processing unit 54 of the trial calculation processing unit 51 performs the insurance money simulation processing according to the trial calculation condition data and the trial calculation instruction from the user terminal 7 (step S5). The trial calculation processing unit 51 may perform the processing in such an order, or may perform only one of the steps.
Further, the processing may be performed by an arbitrary combination of the three steps. Further, the processing may be performed in another order instead of this order.

Next, the details of the distribution scheme change processing by the distribution scheme change trial calculation processing unit 52 and the like will be described with reference to FIGS. 3 to 6. The trial calculation processing unit 5 receives the trial calculation condition data from the user terminal 7, and stores it in the trial calculation condition data storage unit 57 (step S11). It is assumed here that the user terminal 7 has instructed at least the execution of the distribution scheme change process. The trial calculation condition data includes information that specifies insurance contracts and claim data to be subject to trial calculation (for example, accident year or contract year or a combination thereof,
Or specific region, insurance type, etc.) and setting information of the reinsurance allocation scheme to be applied and, if necessary, information of the pattern code setting table. Regarding the setting information of the reinsurance distribution scheme and the information of the pattern code setting table to be applied, if the past information stored in the trial calculation data storage unit 56 is designated by the user terminal 7, it may be used. is there.

Next, the distribution scheme change trial calculation processing unit 52
Specifies the pair of insurance contract and claim data which is the trial calculation target in the trial calculation data storage unit 56 by using the information for identifying the trial calculation target insurance contract and the complaint data included in the trial calculation condition data (step S13). . Then, the present reinsurance allocation scheme is specified for the pair of insurance contract and claim to be calculated, which are specified in step S13, and the reinsurance premium and the recovery insurance when the specified reinsurance allocation scheme is applied. The money and the like are calculated and stored in the storage device (step S15).

To specify the reinsurance allocation scheme,
For example, the order is as shown in FIG. Using the contents of the insurance contract data 100 stored in the trial calculation data storage unit 56, the pattern code setting table 101 is referred to and the distribution scheme code 102 is acquired. Then, using the allocation scheme code 102, the scheme content table 10
3, the contents of the reinsurance allocation scheme 104 are acquired.

The pattern code setting table 101 is a table as shown in FIG. 5, for example. That is, a pattern code column, a condition column, and a distribution scheme code column are provided. Conditions corresponding to one pattern code are listed in the condition column. For example, it is specified by the insurance item, region, property name, and the like. However, it is also possible to specify it under other conditions. Then, the distribution scheme code corresponding to such a pattern code is stored in the column of the distribution scheme code. The allocation scheme code may be the same even if the pattern code and conditions are different.

The scheme content table 103 associates the allocation scheme code with the content of the reinsurance allocation scheme. An example of a reinsurance allocation scheme is shown in FIG. In this example, of the insurance amount of 1 billion yen, up to 100 million yen actually insured is owned by the primary insurance company (all burdened), and more than 100 million yen actually generated up to 1 billion yen 50% of the insurance money of the
0% is reinsurance for other insurance companies. The contents of the reinsurance allocation scheme include information on the reinsurance form regarding such reinsurance and holding, as well as information on the insurance company (cedent) who underwrites the reinsurance and information on the reinsurance premium (actual reinsurance). The charge amount or calculation method) is also included. In some cases, information on the insurance portion of the owned portion may be retained. The reinsurance allocation scheme can be configured arbitrarily, and can be specified in various ways, such as a case where a certain percentage of the insurance amount is ceded, a case where a part or all of the portion exceeding the certain amount is ceded.

When the method for specifying the reinsurance allocation scheme shown in FIG. 4 is used, the reinsurance allocation scheme is changed by changing either or both of the pattern code setting table 101 and the scheme content table 103. be able to. However, the method of identifying the reinsurance allocation scheme shown in FIG. 4 is an example, and the insurance contract data 100
The table may be such that the content of the reinsurance allocation scheme 104 can be directly obtained from the table or the logic for determination can be prepared and specified.

If the reinsurance allocation scheme to be applied in step S15 is specified and the incurred insurance money included in the claim data corresponding to the insurance contract data is applied to the reinsurance allocation scheme, the recovery insurance money and the reinsurance premium are collected. Can be calculated. In addition, if the collected insurance money is subtracted from the insurance money incurred, the net insurance money borne by the insurance company of the underwriter will be calculated. As for the insurance premium, the net insurance premium (= insurance received by the primary insurance company-reinsurance premium) may be calculated.

Next, the distribution scheme change trial calculation processing unit 52
Outputs the trial calculation result and the like to the trial calculation result storage unit 58 for each contract unit, reinsurance reinsurance destination (insurance company that underwrites reinsurance), and the like (step S17). It is also possible to output information about the simple total amount of collected insurance money, the total amount of reinsurance premiums, and the total net insurance money. The user terminal 7 that has requested the trial calculation process via the network 2
May be sent directly to. However, since the trial calculation process often takes a long time, the trial calculation result storage unit 58 once outputs the trial calculation result storage unit 58, and when the user terminal 7 accesses to obtain the trial calculation result, the trial calculation processing unit 51 specifies from the trial calculation result storage unit 58. The result of trial calculation is read, necessary aggregation processing is performed, and the result is output to the user terminal 7.

As described above, by applying a reinsurance allocation scheme different from the actual one to the pair of the insurance contract data and the actually generated claim data and comparing and evaluating, for example, the amount of collected insurance money, It will be possible to assess the effect of reinsurance in the situation in which it occurred.

Next, the trial calculation processing unit 5 will be described with reference to FIGS.
The details of the random trial calculation processing by the first random trial calculation processing unit 52 and the like will be described. The trial calculation processing unit 51 receives the trial calculation condition data from the user terminal 7 and stores it in the trial calculation condition data storage unit 57 (step S21). Here, it is assumed that the user terminal 7 has instructed at least the execution of the random trial calculation process. The trial calculation condition data includes information such as the type of insurance and the trial calculation data year for specifying the trial calculation object, information about the reinsurance allocation scheme, category information for categorizing the trial calculation object, trial calculation processing stop conditions such as the number of trial calculations. Contains information about. The trial calculation condition data may not be received at this stage, but the information designated in step S11 of FIG. 3 may be used as it is. In addition, as information for specifying the target of trial calculation, it is possible to specify insurance contracts and claim data relating to specific disasters. That is, the trial calculation data storage unit 56 stores a data set regarding a specific disaster individually or in a searchable manner.

Next, the random trial calculation processing unit 52 uses the insurance item data for specifying the trial calculation object included in the trial calculation condition data and the information such as the trial calculation data year, and the insurance contract data and the claim data to be the trial calculation target. Is specified in the trial calculation data storage unit 56 (step S22). Also,
The random trial calculation processing unit 53 inspects each claim data that is the trial calculation target, collects the claim data of the same accident number, totals the insured insurance money, generates one claim data, and stores it in the storage device. Store (step S23).
Insurance payments may be divided into multiple payments, and unpaid payments may also be recorded. Therefore, the claims data accrued for paid and unpaid claims data are aggregated and collected. Including incurred insurance money 1
The following processing shall be performed as one complaint data.

At this stage, the insurance contract data and the complaint data due to the accident actually occurred are associated with each other. FIG. 8 shows a schematic diagram showing an example thereof. In this example, the group of insurance contracts is arranged on the left side and the group of claims is arranged on the right side. The insurance contract a2 is associated with the claim c2. Also, claim c
The insurance contract a1 is associated with 1. Claim c
An insurance contract a5 is associated with 4 and c3. As described above, a plurality of claims may correspond to one insurance contract. There are also insurance contracts (for example, a3, a6, and a4) to which no claims are associated. The insurance contract data of the insurance contract a1 includes the security number (a1), item (fire), type (general), property name (plant), job code (chemical industry), location (Tokyo), contractor ( Conspicuous), contract start (April 01, 1999), contract end (March 31, 2000), allocation standard amount (insurance amount:
100 million yen) etc. are included. On the other hand, the claim data of claim c1 includes accident number (c1), accident date (09/28/99), securities number (a1), item (fire), type (general), property name (factory). , Job code (chemical industry), location (Tokyo), contractor (conspicuous), insured insurance (50 million yen), etc.

Then, the random trial calculation processing unit 52, according to the category information designated in the trial calculation condition data,
A categorization process is performed (step S25). This categorization processing classifies the insurance contract data and the claim data, which are the subject of trial calculation, into categories. Items used for category classification include, for example, property type, location, job code, insurance amount, mutual insurance category, mutual insurance share, and the like. For example, when classifying by type of factory, house, office building,
When the location is divided by prefecture, by job code, when the insurance amount is divided by a predetermined insurance amount range, and when the mutual insurance category is specified by a certain condition (single contract of user's insurance company only / user Contract under which the insurance company of the company is the secretary (only the insurance company of the user is subject to calculation) /
When divided into categories that satisfy contracts under management of other companies (only for users' insurance companies), the insurance amount of the user's insurance company relative to the total insurance amount of the mutual insurance contract (%) ) And so on. Note that the configuration may be such that these items are combined into categories. The classification is performed in this manner because the claim and the insurance contract are rearranged in step S27 described below, but in that case, a combination that cannot actually occur, such as insurance claims> insurance amount, does not occur, or This is to maintain a certain reasonable correlation between the claim and the contract pair. That is, the categories are intended to limit the combinations of insurance contracts and claims that may actually occur.

For example, FIG. 9 shows an example in which the example of FIG. 8 is classified into locations. In FIG. 9, the part surrounded by the dotted line is each category. That is, the category of Tokyo, the category of Osaka, the category of Nagoya, and the category of Hokkaido are shown. At this stage, claims and insurance contracts due to an accident that has actually occurred are associated with each other, and claim c1 and insurance contract a1 and claim c2 and insurance contract a2 are associated with each other.

Then, the random trial calculation processing unit 52 rearranges claims and insurance contracts within each category by using random numbers such as pseudo-random numbers and quasi-random numbers (step S2).
7). For example, a random number is generated for one claim data and one insurance contract data specified by the random number is associated. At this time, the same insurance contract data may be associated with a plurality of claim data. By using quasi-random numbers, it is possible to accelerate the convergence of the expected value of the total recovery insurance money, which will be described later. However, the random number generation processing amount may be larger than the pseudo random number. By doing so, the claim data due to an accident that has actually occurred is associated with another insurance contract data in which a similar accident may occur (may be associated with the same insurance contract data). In other words, an accident that actually occurred will occur in another insurance contract.

When the rearrangement is performed in step S27 when the categories are classified as shown in FIG. 9, the result is, for example, as shown in FIG. That is, the insurance contract a1 is actually
The claim c1 associated with is associated with the insurance contract a4. The claim c2 associated with the insurance contract a2 is associated with the insurance contract a3.
This is a rearrangement in the Tokyo area category,
Reclassification is also performed in categories such as Osaka, Nagoya, and Hokkaido. The rearrangement result is held in the storage device of the trial calculation server 5 in a table format as shown in FIG. 11, for example. The table of FIG. 11 is provided with a column of the number of trial calculations, a column of regions representing categories, a column of securities numbers, and a column of accident numbers. In this way, the correspondence between insurance contract data and complaint data is represented by the security number and accident number. It should be noted that, as will be described later, since the rearrangement is performed every time the collected insurance money is calculated for all the claim data, the rearrangement result is stored in the table for each time.

Next, the random trial calculation processing unit 53 acquires the reinsurance allocation scheme applied to the insurance contract data corresponding to the claim data based on the result of the rearrangement in step S27, and calculates the recovery insurance amount and the like. (Step S2
9). The reinsurance allocation scheme is acquired by the processing shown in FIG. 4, for example. For example, if there is claim data as shown in FIG. 8, and the reinsurance allocation scheme is as shown in FIG.
In the case of 2 (a), the recovery insurance money as shown in FIG. 12 (b) is calculated. That is, the insurance company pays up to 30 million yen, which is 30% of the insurance amount of 100 million yen (distribution standard amount). On the other hand, the net insurance premium received by the primary insurance company is 3 million yen, which is 10% of the insurance amount. In addition, in the special provision 1, the insurance company that underwrites the reinsurance is 20% of the insurance amount of 100 million yen with a ceding ratio of 20%.
Pay up to, 000,000 yen. On the other hand, the reinsurance premium received by the insurance company that underwrites the reinsurance is 20%, which is 10% of the ceded reinsurance amount.
It's 0,000 yen. In the special provision 2, the insurance company that underwrites the reinsurance is 3% with a ceding ratio of 30% of the insurance amount of 100 million yen.
Pay up to 10 million yen. On the other hand, the reinsurance premium received by the insurance company underwriting the reinsurance is 10% of the ceded reinsurance amount of 3%.
It is, 000,000 yen. Under special provision 3, the insurance company that underwrites the reinsurance will bear up to 20 million yen, which is a 20% reinsurance rate out of the 100 million yen insured amount. On the other hand, the reinsurance premium received by the insurance company that underwrites the reinsurance is 10% of the ceded reinsurance amount, which is 2 million yen. Note that, as in the example of FIG.
When it is determined that the insurance premium is 10% of the insurance premium, for example, the insurance premium may be calculated in step S29.

As shown in FIG. 12 (b), when the insured insurance money is 50 million yen, the direct insurance company has 30
Bear 15 million yen of insurance money. The insurance companies that underwent reinsurance for Special Contract 1 had a ceding ratio of 20%, 10
Bear, 000,000 yen. The insurance company that underwent reinsurance for Special Contract 2 will bear 15 million yen, which is a 30% ceding ratio. The insurance company that underwrote reinsurance for Special Contract 3 will bear 10 million yen, which is a 20% ceding ratio. In this example, the net insurance money borne by the primary insurance company is 15 million yen, and 35 million yen is the collected insurance money.

In step S29, the random trial calculation processing unit 53 calculates the recovery insurance money for all the complaint data. Then, for example, a table including the trial calculation results as shown in FIG. 13 is stored in the trial calculation result storage unit 58. In the example of FIG. 13, a column of the number of trial calculations, a column of regions representing categories
Security number column, accident number column, insurance amount (primary insurance amount) column, pattern code column, incurred insurance amount column,
It includes a net insurance money column and a recovery insurance money column. Also,
As shown in FIG. 14, information about each covenant is also stored in the trial calculation result storage unit 58 corresponding to each row. That is,
For each covenant, the ceding ratio and the amount of insurance claims collected are recorded.

It should be noted that the total recovery insurance amount is calculated by summing the recovery insurance amounts in the table of FIG. 13 in the same trial calculation times. Then, the expected value for the total collected insurance money is further calculated and stored in the storage device (step S31). It is sufficient to add up the total insurance claims calculated so far and divide by the number of trial calculations.

Then, the random trial calculation processing unit 53 determines whether or not the stop condition of the rearrangement processing (trial calculation processing) is satisfied (step S33). For example, when the trial calculation count is directly specified in the trial calculation condition data, it is determined whether the trial calculation count has been reached. On the other hand, if, for example, the fluctuation range of the expected value of the total collected insurance money is specified instead of the number of times of trial calculation, the expected value of the total collected insurance money calculated in the previous step S31 and the total collected amount calculated this time in step S31 The difference from the expected value of the insurance money is calculated, and it is judged whether or not it is within the deviation range which is the condition for stopping the recombination process. If the condition for stopping the rearrangement process is not satisfied, step S
Return to 27.

On the other hand, when it is judged that the stop condition of the rearrangement process is satisfied, various totalization processes are executed (step S35). The totaling process performed in step S35 may be specified in, for example, trial calculation condition data, or may be predetermined. Here, for example, a ranking list that indicates whether the trial calculation is the largest or the smallest total collected insurance money, detailed data in the ranking list (contract unit, further contract unit), collection by pattern code In step S35, information such as the insurance money, the collectable insurance money by special contract, and the information of the convergence distribution graph showing the convergence status of the expected value of the total collected insurance money is generated. When the table as shown in FIG. 13 is used, the collection of collected insurance money for each pattern code can be performed very easily. Then, the random trial calculation processing unit 53 stores the processing result of step S35 in the trial calculation result storage unit 58 (step S37).

The user operates the user terminal 7 at an appropriate time to access the trial calculation server 5 and obtain the result of the designated trial calculation processing. That is, the trial calculation server 5, which has received the request for obtaining the result of the trial calculation process from the user terminal 7, reads the result of the trial calculation process instructed by the user from the trial calculation result storage unit 58 and transmits the result to the user terminal 7. The user terminal 7 that has received the result of the trial calculation process displays the trial calculation result on the display device.

By carrying out such processing, an accident that actually occurs is caused to occur in another insurance contract in which a similar accident may occur, and the recovery insurance money at that time is designated as a reinsurance. By calculating according to the allocation scheme, it becomes possible to see the power of reinsurance. Also, by looking at the expected value of the total amount of insurance claims collected, it becomes possible to make an appropriate evaluation of reinsurance. However, since it is calculated based on actual accidents, accidents that have not occurred in the past cannot be considered.

Next, the processing flow of insurance money simulation processing by the insurance money simulation processing unit 54 and the like will be described with reference to FIG. The trial calculation processing unit 51 receives the trial calculation condition data from the user terminal 7 and stores it in the trial calculation condition data storage unit 57 (step S41). It is assumed here that the user terminal 7 has instructed at least the insurance money simulation process. The trial calculation condition data includes information for specifying the trial calculation target, category information, information about the reinsurance allocation scheme, trial calculation frequency, and the like. The trial calculation condition data may not be received at this stage, but the information designated in step S11 of FIG. 3 may be used as it is.

Next, the insurance money simulation processing unit 54
Is the distribution of the frequency of occurrence of the most suitable accident (for example, Poisson distribution) based on the past complaint data (claim data within the range specified as the trial calculation target) stored in the trial calculation data storage unit 56. ) Is determined and stored (step S43). For example, the maximum likelihood estimation method is used to determine the parameter, and the parameter is λ in the case of Poisson distribution. In addition, based on past complaint information, the average, variance, and distribution of the expected loss amount of the most suitable accident, etc. (in this case, including the loss amount or insurance claims, etc., all of the indices showing the size of the damage) A parameter value representing (for example, Pareto distribution or lognormal distribution) is determined and stored (step S44). In the case of the Pareto distribution, the parameters are a and b, in the case of the log-normal distribution, the parameters are μ and σ, and the maximum likelihood estimation method is used to determine the parameters.

The distribution of the occurrence frequency of accidents, for example, is as shown in FIG. 16, for example. In FIG. 16, the horizontal axis represents the number of annual accidents and the like, and the vertical axis represents the probability of occurrence. The distribution of the expected damage amount such as an accident is as shown in FIG. 17, for example. In FIG. 17, the horizontal axis is the damage amount, and the vertical axis is the occurrence probability. Regarding the parameter values,
It may be presented to the user for correction.

Then, using the distribution of the frequency of occurrence of accidents and the probability of occurrence caused by random numbers, which is defined by the determined parameter values, the expected number of complaints that will occur in a certain year is determined and stored in the storage device. (Step S45).
In addition, using the distribution of expected damages such as accidents and the probability of occurrence caused by random numbers, which are defined by the determined parameter values, the expected damages caused for each expected complaint are generated and stored in the storage device. (Step S46). In this way, expected claim data for a given year is generated. Information on frequency of occurrence (expected number of complaints), information on expected damage amount, and expected complaint data,
It may be generated for each category used in the random trial calculation process (for example, insurance target property, location, insurance amount, other job work, etc.). If the expected damage amount exceeds the insurance amount, the expected damage amount is generated again using a random number.

Next, the insurance money simulation processing unit 54
Associates each expected complaint data with an insurance contract by using a random number (step S47). In this way, the insurance contract for which each expected complaint data representing an accident or the like is generated is specified. Then, the reinsurance allocation scheme applied to the insurance contract associated with each expected claim data is
Based on the information about the reinsurance allocation scheme included in the trial calculation condition data received in step S41, for example, a method as shown in FIG. 4 is determined, and the recovery insurance amount for each expected claim is calculated according to the reinsurance allocation scheme, The data is stored in the storage device (step S49). And
The total collected insurance money, which is the total of the collected insurance money, is calculated and stored in the storage device.

The insurance simulation processing section 54 further calculates the expected value for the total collected insurance money, and stores it in the storage device (step S47). It is sufficient to add up the total insurance claims calculated so far and divide by the number of trial calculations.

Then, the insurance simulation processing unit 54
Determines whether the trial calculation stop condition is satisfied (step S
49). For example, the number of trial calculations (for example, 1
(10,000 times) is directly specified, it is judged whether the trial calculation count has been reached. On the other hand, if, for example, the fluctuation range of the expected value of the total collected insurance money is specified instead of the number of trial calculations, the expected value of the total collected insurance money calculated in the previous step S49 and the total collected amount calculated in the current step S49 are calculated. The difference from the expected value of the insurance money is calculated, and it is judged whether or not it falls within the fluctuation range that is the condition for stopping the recombination process. If the condition for stopping the rearrangement process is not satisfied, step S
Return to 45.

If it is determined that the trial calculation stop condition is satisfied, various tabulation processing is executed (step S51). The totaling process performed in step S51 may be specified in trial calculation condition data, or may be predetermined. Here, for example, a ranking list that indicates whether the trial calculation is the largest or the smallest total collected insurance money, detailed data in the ranking list (contract unit, further contract unit), collection by pattern code In step S51, information such as the insurance money, the collectable insurance money by special contract, and the information of the convergence distribution graph showing the convergence status of the expected value of the total collected insurance money is generated. Then, the insurance simulation processing unit 54 stores the processing result of step S51 in the trial calculation result storage unit 58 (step S53).

The user operates the user terminal 7 at an appropriate time to access the trial calculation server 5 to obtain the result of the designated trial calculation processing. That is, the trial calculation server 5, which has received the request for obtaining the result of the trial calculation process from the user terminal 7, reads the result of the trial calculation process instructed by the user from the trial calculation result storage unit 58 and transmits the result to the user terminal 7. The user terminal 7 that has received the result of the trial calculation process displays the trial calculation result on the display device.

The process flow described with reference to FIG. 15 is an example, and the present invention is not limited to this. In the above flow, a configuration is shown in which the expected number of claims for a certain year is determined by a random number, but the expected number of claims for each year is generated at an arbitrary scale such as 10,000 years. It may be configured to process the number. Further, the expected damage amount may be first generated from the distribution of the expected damage amount, and then the expected number of claims may be generated by random numbers according to the distribution of the occurrence frequency to generate the expected damage amount. Furthermore, instead of associating the expected claims expected to occur with each insurance contract, each expected claim may be associated with the insurance contract.

By performing such processing, not only the accident that actually occurred but also the accident that may occur although not actually occurred can be calculated, and the recovery insurance money in that case can be calculated. Like Therefore, the reinsurance can be evaluated more appropriately. In addition,
It may be argued that it is virtually different from the reality because an accident occurs virtually, but the trial calculation result of the random trial calculation processing unit 53 and the distribution scheme change trial calculation processing unit 5
It is also possible to further evaluate the deviation from the reality by comparing with the trial calculation based on the actual accident or the like based on the trial calculation result of 2.

The scenes where the user uses this system are as follows, for example. (1) Verification of reinsurance scheme revision effect For example, after settlement of accounts, a trial calculation may be performed for negotiations for reinsurance contract renewal. This is because, based on the results of the trial calculation, the conditions for the special reinsurance contract can be set advantageously. In addition, after changing the reinsurance scheme (the insurance company's holding rules), a trial calculation may be performed in order to evaluate the performance of the recovered insurance money by the changed reinsurance scheme. In addition, if the reinsurance scheme was changed in the previous fiscal year, the reinsurance scheme may be implemented to compare and evaluate the insurance premiums and claims collected in the previous year and the previous year. (2) Earnings forecast by comparison with the previous year This earnings forecast will be made from time to time. When the reinsurance scheme for the current year is renewed, the insurance premiums and collected claims for the previous year and the current year may be compared to predict the collected claims for the current year. (3) Comparison with the holding theory model The comparison with the holding theory model is performed irregularly. Note that the holding theory system is a system that estimates insurance claims by inducing a hypothetical disaster (typhoon, fire, etc.) based on the portfolio of the user's insurance company, and makes a trial calculation to plan and evaluate the reinsurance scheme. Is to do. The retention theory model shows a reinsurance scheme simulated with a retention theory system. For example, in some cases, a reinsurance scheme (holding theory model) evaluated by this holding theory system is used to perform insurance claim calculation using this system, and the claims collection results are compared and evaluated. As a result, it becomes possible to widely verify the effect of insurance money recovery. (4) Verification of scheme effects during large-scale disasters Conduct irregularity, for example, when a large-scale disaster occurs. Using the past production scheme (actually used scheme), the production scheme that was applied when the disaster occurred, the test scheme, and using the claim data at the time of a large-scale disaster, insurance premiums
This is to calculate, compare and evaluate the collected insurance money. Note that the above is an example, and trial calculation may be performed in other situations.

Although one embodiment of the present invention has been described above, the present invention is not limited to this. That is, the system outline diagram shown in FIG. 1 is an example, and the calculation center 3 may be provided with the trial calculation server 5. Also,
The function of the trial calculation server 5 may be provided in the user terminal 7.

[0069]

As described above, according to the present invention, it is possible to provide a technique for making it possible to more appropriately evaluate the ability of reinsurance.

[Brief description of drawings]

FIG. 1 is a schematic diagram of a system according to an embodiment of the present invention.

FIG. 2 is a processing flow showing an outline of processing of trial calculation processing.

FIG. 3 is a diagram showing a processing flow of distribution scheme change trial calculation processing.

FIG. 4 is a schematic diagram for explaining a method of acquiring information on a reinsurance allocation scheme.

FIG. 5 is a diagram showing an example of a pattern code setting table.

FIG. 6 is a diagram showing an example of a reinsurance allocation scheme.

FIG. 7 is a diagram showing a processing flow of random trial calculation processing.

FIG. 8 is a schematic diagram showing a correspondence relationship between insurance contracts and claims.

FIG. 9 is a schematic diagram for explaining categorization.

FIG. 10 is a schematic diagram for explaining rearrangement of insurance contracts and claims.

FIG. 11 is a diagram showing an example of a table that stores a rearrangement result.

FIG. 12 (a) shows an example of a reinsurance allocation scheme,
(B) is a figure which shows the ceding state of the insurance money which occurred.

FIG. 13 is a diagram showing an example of a table that stores calculation results of collected insurance money and the like.

FIG. 14 is a table showing the ceding reinsurance rate and the collected insurance money for each special contract in FIG.

FIG. 15 is a diagram showing a processing flow of insurance money simulation processing.

FIG. 16 is a diagram showing an example of a distribution of occurrence frequencies.

FIG. 17 is a diagram showing an example of distribution of expected damage amount.

[Explanation of symbols]

1, 2 Network 3 Calculation center 5 Trial calculation server 7 User terminal 31 Normal system 33 Reinsurance DB 34 Insurance contract DB 35 Claim DB 37 Trial calculation data creation / transmission server 39 Trial calculation data storage unit 51 Trial calculation processing unit 52 Distribution scheme change trial calculation Processing unit 53 Random trial calculation processing unit 54 Insurance money simulation processing unit 55 Download processing unit 56 Trial calculation data storage unit 57 Trial calculation condition data storage unit 58 Trial calculation result storage unit

Continued front page    (72) Inventor Takeshi Itoh             1-2-1, Marunouchi, Chiyoda-ku, Tokyo East             Kyokai Fire Insurance Co., Ltd. (72) Inventor Tadao Furuya             890 Kashimada, Sachi-ku, Kawasaki City, Kanagawa Stock             Hitachi, Ltd. Financial Systems Division (72) Inventor Shigeru Yamauchi             890 Kashimada, Sachi-ku, Kawasaki City, Kanagawa Stock             Hitachi, Ltd. Financial Systems Division

Claims (19)

[Claims] 1. Claimed claim data for a claim that has actually occurred is associated with arbitrary insurance contract data,
The association step of storing the correspondence relationship in the storage device, and the reinsurance allocation scheme applied to the insurance contract data associated in the associating step with each of the generated claim data are determined, A reinsurance claim recovery amount calculation step for calculating a reinsurance claim recovery amount for each of the claims according to a reinsurance allocation scheme, and storing the reinsurance claim recovery amount in a storage device. 2. The reinsurance evaluation method according to claim 1, wherein in the associating step, the generated complaint data is associated with insurance contract data based on a random number. 3. The method further comprises a step of categorizing the insurance contract data to be processed and the generated claim data based on category information designated as a trial calculation condition, and the associating step includes, for each category, a corresponding category The reinsurance evaluation according to claim 1 or 2, further comprising a step of associating each of the included claim data with any insurance contract data included in the category and storing the correspondence in a storage device. Method. 4. The category information is specified by at least one of an insurance target property, a location, an insurance amount, job work, and in the case of joint insurance, the contents of the joint insurance. Reinsurance evaluation method described in 3. 5. A step of repeatedly performing the associating step and the reinsurance claim recovery amount calculation step, calculating an expected value of the total reinsurance claim recovery amount for all claims in each repetition, and storing the expected value in a storage device. The reinsurance evaluation method according to any one of claims 1 to 4, further comprising: 6. The method according to claim 1, further comprising a step of totalizing the reinsurance recovery amount for each claim calculated in the reinsurance recovery amount calculating step for each insurance contract pattern for determining a reinsurance allocation scheme. The reinsurance evaluation method described in any one of 5. 7. A step of calculating a reinsurance premium based on the reinsurance distribution scheme applied to the insurance contract data associated with each of the generated claim data in the associating step, and storing the reinsurance premium in a storage device. The reinsurance evaluation method according to any one of claims 1 to 6, further comprising: 8. In the reinsurance claim recovery amount calculating step, a reinsurance allocation scheme applied to the insurance contract data associated with each of the generated claim data in the associating step is set as an arbitrary reinsurance. The reinsurance evaluation method according to any one of claims 1 to 7, wherein the reinsurance evaluation method is performed by using input information regarding an allocation scheme. 9. The specified reinsurance allocation scheme, wherein the reinsurance allocation scheme applied to the actual insurance contract data corresponding to the actually generated claim data is changed to a specified reinsurance allocation scheme. 9. The reinsurance evaluation method according to any one of claims 1 to 8, further comprising a step of calculating a reinsurance claim recovery amount according to each claim and storing it in a storage device. 10. A step of storing input information relating to an arbitrary reinsurance allocation scheme in a storage device, and a reinsurance allocation applied to actual insurance contract data corresponding to accrued claim data for an actually generated claim. Determining a scheme using the input information, calculating a reinsurance recovery amount for each claim according to the determined reinsurance allocation scheme, and storing the reinsurance recovery amount in a storage device. 11. A frequency / damage amount generation step of generating frequency of occurrence and expected damage amount data based on information on actually occurring complaints and storing the data in a storage device, and based on the occurrence frequency and expected damage amount data. Generating expected claim data, generating expected claim data relating to each expected claim, and storing each said expected claim data in a storage device; and corresponding said expected claim data to insurance contract data And a correspondence step of storing the correspondence relationship in a storage device, a distribution scheme storage step of storing input information regarding an arbitrary reinsurance distribution scheme in the storage device, and the correspondence step for each of the expected claim data. Enter the reinsurance allocation scheme applied to the insurance contract data associated in Make decisions using information,
A reinsurance recovery amount calculation step of calculating a reinsurance recovery amount for each of the expected claims according to the determined reinsurance allocation scheme and storing the reinsurance recovery amount in a storage device. 12. The frequency / expected damage amount data is generated in the frequency / damage amount generation step by setting a frequency distribution and an expected damage amount distribution based on information about an actually generated complaint. 12. The reinsurance evaluation method according to claim 11. 13. The reinsurance according to claim 11, wherein, in the expected claim data generating step, the expected claim is generated based on a random number by using the occurrence frequency and the expected damage amount data. Evaluation methods. 14. A reinsurance allocation scheme storing step of storing input information relating to an arbitrary reinsurance allocation scheme in a storage device, and generated claim data for an actually generated claim in association with arbitrary insurance contract data, Calculating a reinsurance recovery amount for each of the claims according to a reinsurance allocation scheme based on the input information, which is applied to insurance contract data associated with each of the incurred claim data, and stores the reinsurance amount in a storage device; 1 Reclaim payment calculation step and expected complaints according to the frequency of occurrence and the expected damage amount generated based on the information about the claims actually occurred.
Generating data and associating the expected claim data with insurance contract data, and applying the reinsurance allocation scheme based on the input information applied to the insurance contract data associated with each of the expected claim data. A second reinsurance claim recovery calculation step of calculating a reinsurance claim recovery amount for the expected claim and storing it in a storage device. 15. The reinsurance evaluation method according to claim 14, further comprising the step of summing the reinsurance claim recovery amounts calculated for each claim in the first and second reinsurance claim recovery amount calculating steps. 16. A program for causing a computer to execute the reinsurance evaluation method according to any one of claims 1 to 15. 17. An associating unit for associating the generated claim data of an actually generated claim with arbitrary insurance contract data and storing the corresponding relationship in a storage device, and for each of the generated claim data. The reinsurance allocation scheme applied to the insurance contract data associated by the associating means is determined, the reinsurance claim recovery amount is calculated for each of the claims in accordance with the determined reinsurance allocation scheme, and stored in the storage device. A computer system having a reinsurance claim recovery amount calculation means. 18. A frequency / damage amount generation means for generating frequency of occurrence and expected damage amount data based on information on actually occurring complaints and storing the data in a storage device, and based on the occurrence frequency and expected damage amount data. Expected claim data is generated by generating predicted claim data for each expected claim, and the expected claim data is stored in a storage device, and the expected claim data is associated with insurance contract data. And a correspondence means for storing the correspondence relationship in a storage device, a storage means for storing input information relating to an arbitrary reinsurance distribution scheme, and an insurance related by the correspondence means to each of the expected claim data. The reinsurance allocation scheme applied to the contract data is determined using the input information and the determined reinsurance allocation ski And a reinsurance recovery amount calculation means for calculating a reinsurance recovery amount for each of the expected claims according to the system and storing the reinsurance recovery amount in a storage device. 19. Corresponding reinsurance allocation scheme designation receiving means for storing input information relating to an arbitrary reinsurance allocation scheme in a storage device, and generated claim data for an actually generated claim to arbitrary insurance contract data. , A reinsurance claim recovery amount is calculated for each of the claims according to a reinsurance allocation scheme based on the input information, which is applied to insurance contract data associated with each of the incurred claim data, and stored in a storage device. Expected complaints according to the first reinsurance claim recovery amount calculation means and information on the frequency of occurrence and the expected damage amount generated based on the information on the claims actually occurred.
Generating data and associating the expected claim data with insurance contract data, and applying the reinsurance allocation scheme based on the input information applied to the insurance contract data associated with each of the expected claim data. A second reinsurance claim recovery amount calculation means for calculating a reinsurance claim recovery amount for the expected claim and storing it in a storage device.