JPH10228462A - Path summarizing method by monte carlo simulation - Google Patents

Path summarizing method by monte carlo simulation

Info

Publication number
JPH10228462A
JPH10228462A JP3166697A JP3166697A JPH10228462A JP H10228462 A JPH10228462 A JP H10228462A JP 3166697 A JP3166697 A JP 3166697A JP 3166697 A JP3166697 A JP 3166697A JP H10228462 A JPH10228462 A JP H10228462A
Authority
JP
Japan
Prior art keywords
scenario
correlation coefficient
probabilistic
file
condition
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP3166697A
Other languages
Japanese (ja)
Inventor
Takashi Matsumura
隆史 松村
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hitachi Ltd
Original Assignee
Hitachi Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hitachi Ltd filed Critical Hitachi Ltd
Priority to JP3166697A priority Critical patent/JPH10228462A/en
Publication of JPH10228462A publication Critical patent/JPH10228462A/en
Pending legal-status Critical Current

Links

Landscapes

  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

PROBLEM TO BE SOLVED: To execute simulation on the same condition in a short time and also to accomplish an equivalent result by integrating similar probabilistic scenarios among plural probabilistic scenario groups that are generated by random numbers by using a correlation coefficient and narrowing them. SOLUTION: A numbering device 12 numbers a scenario number to a probabilistic scenario that is generated by using random number stored in a scenario file 11, and the numbered probabilistic scenario is stored in a numbering file 13. A summarization condition setting device 14 designates a correlation coefficient as a condition to summarize the probabilistic scenario and sets a condition that a scenario belongs to the same group when a correlation coefficient is a designated condition or more. Next, the device 14 compares whether a correlation coefficient that is calculated by a summarizing device 15 exists within the range of a correlation coefficient that is optionally set by a user and stores it in a summarization file 16.

Description

【発明の詳細な説明】DETAILED DESCRIPTION OF THE INVENTION

【0001】[0001]

【発明の属する技術分野】本発明は確率論的シナリオを
基にシミュレーションを行うことによって保険料等を決
定する方法に関する。
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for determining insurance premiums and the like by performing a simulation based on a stochastic scenario.

【0002】[0002]

【従来の技術】保険料を決定する際には、金利や死亡
率、人件費や間接費といった事業費などを将来数十年に
渡って設定し、これらの数値をもとに将来の収入・支出
などの予測である将来収支予測(キャシュフローテス
ト)を行ったり、各種保険料の計算を行ったりする必要
がある。これらの数値は、一定、逓増、逓減といったよ
うに、人が任意に設定することも可能であり、この決定
されたシナリオは、決定論的シナリオと呼ばれる。
2. Description of the Related Art When determining insurance premiums, business expenses such as interest rates, mortality rates, labor costs and indirect costs are set over the next several decades, and future revenue and income are determined based on these figures. It is necessary to perform a future income and expenditure forecast (cash flow test), which is a forecast of expenditures, and to calculate various insurance premiums. These numerical values can be arbitrarily set by a person, such as constant, increasing, decreasing, and the like, and the determined scenario is called a deterministic scenario.

【0003】決定論的シナリオでは、実際の金利、死亡
率などのような不規則な変化を設定することが困難なた
め、乱数をもとに不規則な変動のシナリオを作成する手
法があり、この手法により作成されたシナリオを確率論
的シナリオと呼ぶ。
In the deterministic scenario, it is difficult to set an irregular change such as an actual interest rate or a mortality rate. Therefore, there is a method of creating an irregular change scenario based on random numbers. The scenario created by this method is called a stochastic scenario.

【0004】将来収支予測を行う、キャッシュフローテ
ストや保険料算出などの計算を実施するシミュレーショ
ンでは、乱数発生装置によって発生された乱数を金利や
死亡率などの変動を表現する計算式に代入し、このよう
にして生成された金利、死亡率等の確率論的シナリオを
もとに保険料算出等の各種計算をn回実施し、算出され
た数値の分布を分析することにより、保険料等を決定す
る。
[0004] In a simulation for performing a calculation such as a cash flow test or an insurance premium calculation for predicting a future income and expenditure, a random number generated by a random number generator is substituted into a calculation expression expressing a change such as an interest rate or a mortality rate. Based on the probabilistic scenarios such as interest rates and mortality rates generated in this way, various calculations such as insurance premiums are performed n times, and the distribution of the calculated numerical values is analyzed, whereby insurance premiums and the like are calculated. decide.

【0005】[0005]

【発明が解決しようとする課題】従来の方法では、パラ
メータの制御、制約条件の設定を行わずに乱数を発生
し、確率論的シナリオを生成しているため、これらのシ
ナリオをもとに保険料等の算出を逐一、実施する場合、
保険料算出等のための処理時間が長時間かかるという問
題があった。
In the conventional method, random numbers are generated without controlling parameters and setting constraints, and probabilistic scenarios are generated. Therefore, insurance is performed based on these scenarios. When calculating the fees etc. one by one,
There is a problem that a long processing time is required for calculating the insurance premium.

【0006】[0006]

【課題を解決するための手段】本発明は、乱数によって
複数発生された確率論的シナリオ群を相関係数を利用し
て、類似した確率論的シナリオを統合し、絞り込む手段
に関する。本発明により同一条件のシミュレーションを
短時間で実施し、かつ同等の成果を上げることが可能に
なる。
SUMMARY OF THE INVENTION The present invention relates to a means for integrating and narrowing down similar stochastic scenarios using a correlation coefficient for a plurality of stochastic scenarios generated by random numbers. According to the present invention, it is possible to perform a simulation under the same conditions in a short time and achieve the same result.

【0007】[0007]

【発明の実施の形態】以下、本発明の実施形態を詳細に
説明する。図1は本発明を適用した場合の処理手順を示
すフローチャートであり、図2は本発明に関わる集約装
置の構成を示すブロック図であり、図3は前提条件とし
て存在する、乱数を利用して生成された、自由金利の変
動の様子をシミュレーションする
DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail. FIG. 1 is a flowchart showing a processing procedure when the present invention is applied, FIG. 2 is a block diagram showing a configuration of an aggregation device according to the present invention, and FIG. 3 uses a random number existing as a precondition. Simulate the generated free interest rate fluctuation

【0008】[0008]

【数1】dr(t)=(1-β)(α-r(t))dt+σ√r(t)ε(t) という平均回帰過程(CIRモテ゛ル)により発生された、集約
される前の確率論的シナリオの格納されているシナリオ
ファイル(11)の内容であり、各シナリオNo毎の時系列
の値の変化を示している。
## EQU00001 ## Generated by a mean regression process (CIR model), dr (t) = (1-.beta.) (. Alpha.-r (t)) dt + .sigma.r (t) .epsilon. (T), aggregated This is the content of the scenario file (11) in which the previous probabilistic scenario is stored, and shows a change in a time-series value for each scenario No.

【0009】(α,βは定数でここでは、α=0.05,β=0.
5としている。また、ε(t)は正規乱数である。) 図4は図3に示されたような、前提条件としてあたえら
れた一般にm本存在する確率論的シナリオの、2本のシ
ナリオ毎の相関係数を示したものである。本実施例にお
いては、10種類の確率論的シナリオを例示し、2本毎の
全ての組合せの相関係数を算出している。図5は集約後
の確率論的シナリオの格納されている集約ファイル(1
6)である。
(Α and β are constants, where α = 0.05 and β = 0.
And 5. Ε (t) is a normal random number. FIG. 4 shows the correlation coefficient of each of two generally existing stochastic scenarios given as preconditions, as shown in FIG. In the present embodiment, ten types of probabilistic scenarios are exemplified, and the correlation coefficients of all combinations of every two are calculated. Figure 5 shows the aggregated file (1
6).

【0010】図1に従い、以下本発明について詳しく説
明する。ファイルとして入力するシナリオファイル(1
1)に格納されているm本の乱数を利用して生成された確
率論的シナリオにシナリオ番号を打番装置(12)により
打番し、打番された確率論的シナリオを打番ファイル
(13)に格納する(100)。
The present invention will be described below in detail with reference to FIG. Scenario file (1
The probabilistic scenario generated using the m random numbers stored in 1) is numbered with a scenario number by a numbering device (12), and the numbered probabilistic scenario is recorded in a numbering file ( Stored in 13) (100).

【0011】次に、集約条件設定装置(14)により、確
率論的シナリオを集約するための条件として、相関係数
を指定し、相関係数が指定した条件以上であれば、同一
グループであるというような条件を設定する。本実施例
では、相関係数として0.5を設定し、0.5以上であれば、
同一グループと見なし、集約する条件を指定する(10
2)。
Next, the aggregation condition setting device (14) specifies a correlation coefficient as a condition for integrating the probabilistic scenarios, and if the correlation coefficient is equal to or more than the specified condition, the groups belong to the same group. Set conditions such as: In this embodiment, 0.5 is set as the correlation coefficient, and if it is 0.5 or more,
Consider the same group and specify the conditions for aggregation (10
2).

【0012】次に、集約装置(15)により打番ファイル
(13)からn本ある確率論的シナリオのうちの2本を選
択し、時間t0からtnまでのシナリオkの集合をa
i、平均をa、シナリオlの集合をbi、平均をb、相
関係数をrとし、
Next, two of the n probabilistic scenarios are selected from the hit number file (13) by the aggregation device (15), and a set of scenarios k from time t0 to tn is defined as a
i, the average is a, the set of scenarios l is bi, the average is b, the correlation coefficient is r,

【0013】[0013]

【数2】r=Σ(ai−a)(bi−b)/√Σ(ai
−a)2・(bi−b)2 の数式に従い、相関係数を算出する(104)。例えば、
シナリオNo1とシナリオNo2との相関係数は、この式に従
って0.38と計算され図4のとおりに格納できる。
R = 2 (ai-a) (bi-b) / a (ai
A correlation coefficient is calculated according to the equation of -a) 2 · (bi-b) 2 (104). For example,
The correlation coefficient between scenario No. 1 and scenario No. 2 is calculated to be 0.38 according to this equation and can be stored as shown in FIG.

【0014】次に、集約装置(15)にて算出された相関
係数が集約条件設定装置(14)で、利用者が任意に設定
した相関係数の範囲(本実施例では0.5)内にあるかど
うかを比較する(106)。
Next, the correlation coefficient calculated by the aggregation device (15) is within the range of the correlation coefficient arbitrarily set by the user (0.5 in this embodiment) by the aggregation condition setting device (14). It is compared whether there is (106).

【0015】相関係数が指定された範囲内であれば、2
本の確率論的シナリオが同一グループであると見なし、
打番装置(12)にて打番ファイル内の確率論的シナリオ
の内、番号の小さな確率論的シナリオを集約ファイルに
格納する。
If the correlation coefficient is within the specified range, 2
Consider the stochastic scenarios in the book to be in the same group,
The hitting device (12) stores a stochastic scenario with a small number among the stochastic scenarios in the hitting file in an aggregated file.

【0016】相関係数が集約条件設定装置(14)にて、
利用者に設定された範囲外であれば、そのまま2つの確
率論的シナリオを集約ファイルに格納する(110)。
The correlation coefficient is calculated by an aggregation condition setting device (14).
If it is out of the range set by the user, the two stochastic scenarios are stored as they are in the aggregated file (110).

【0017】この一連の(104〜110)の動作を全てのシ
ナリオの組合せにてm(m−1)/2回行う。
This series of operations (104 to 110) is performed m (m-1) / 2 times in all combinations of scenarios.

【0018】本実施例においては、図4に示すように打
番されたシナリオ番号がNo1とNo4、No2とNo4、No4とNo9
の相関係数が0.5以上となっているので、No1とNo4を同
一グループと見なしNo4を集約し、No2とNo4を同一グル
ープと見なしNo4を集約し、No4とNo9を同一グループと
見なしNo9を集約する。
In this embodiment, the scenario numbers numbered as shown in FIG. 4 are No.1 and No.4, No.2 and No.4, No.4 and No.9.
No.1 and No4 are regarded as the same group and No4 is aggregated, No2 and No4 are regarded as the same group, No4 is aggregated, and No4 and No9 are regarded as the same group and No9 is aggregated since the correlation coefficient is 0.5 or more I do.

【0019】集約ファイルの中身は、図5のように集約
された確率論的シナリオが残る。
The contents of the aggregated file remain stochastic scenarios aggregated as shown in FIG.

【0020】[0020]

【発明の効果】乱数を利用して発生した確率論的シナリ
オのグループ分けを行い、集約したファイルを作成し、
これを利用することによって、同一条件のシミュレーシ
ョンを集約しない場合より短時間にて実施し、同程度の
効果を得ることが可能となる。
According to the present invention, a group of probabilistic scenarios generated using random numbers is created, and an aggregated file is created.
By using this, it is possible to execute the simulation in a shorter time than when the simulations under the same conditions are not aggregated, and to obtain the same effect.

【図面の簡単な説明】[Brief description of the drawings]

【図1】本発明の処理手順の実施形態を示すフローチャ
ートである。
FIG. 1 is a flowchart showing an embodiment of a processing procedure of the present invention.

【図2】本発明に関わる確率論的シナリオを集約するた
めの装置のシステムブロック図である。
FIG. 2 is a system block diagram of an apparatus for aggregating stochastic scenarios according to the present invention.

【図3】本発明に関わる打番ファイルの構成図である。FIG. 3 is a configuration diagram of a hit number file according to the present invention.

【図4】本発明に関わる集約条件の相関係数を示した図
である。
FIG. 4 is a diagram showing a correlation coefficient of an aggregation condition according to the present invention.

【図5】本発明に関わる集約ファイルの構成図である。FIG. 5 is a configuration diagram of an aggregation file according to the present invention.

【符号の説明】[Explanation of symbols]

10…シナリオ集約装置、 11…シナリオファイル、
12…打番装置、13…打番ファイル、 14…集約
条件設定装置、 15…集約装置、16…集約ファイル。
10… Scenario aggregation device, 11… Scenario file,
12 ... hitting device, 13 ... hitting file, 14 ... consolidation condition setting device, 15 ... consolidation device, 16 ... consolidation file.

Claims (2)

【特許請求の範囲】[Claims] 【請求項1】保険または、年金において確率論的に保険
料算出を実施する方法において、保険料算出のために必
要な乱数を利用することによって生成した、確率論的シ
ナリオを事前に集約し、代表的な確率論的シナリオであ
る代表パスを作成する処理を有することを特徴とするシ
ミュレーションに利用するためのパスの集約方法。
1. A method for stochastically calculating premiums in insurance or pensions, in which stochastic scenarios generated by using random numbers required for calculating premiums are aggregated in advance, A method of aggregating paths for use in a simulation, comprising a process of creating a representative path, which is a representative stochastic scenario.
【請求項2】請求項1の集約処理において、集約の度合
を利用者が設定する機能を有する事を特徴とするシミュ
レーションに利用するためのパスの集約方法。
2. A method according to claim 1, further comprising the step of setting a degree of aggregation by a user.
JP3166697A 1997-02-17 1997-02-17 Path summarizing method by monte carlo simulation Pending JPH10228462A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP3166697A JPH10228462A (en) 1997-02-17 1997-02-17 Path summarizing method by monte carlo simulation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP3166697A JPH10228462A (en) 1997-02-17 1997-02-17 Path summarizing method by monte carlo simulation

Publications (1)

Publication Number Publication Date
JPH10228462A true JPH10228462A (en) 1998-08-25

Family

ID=12337465

Family Applications (1)

Application Number Title Priority Date Filing Date
JP3166697A Pending JPH10228462A (en) 1997-02-17 1997-02-17 Path summarizing method by monte carlo simulation

Country Status (1)

Country Link
JP (1) JPH10228462A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004133491A (en) * 2002-08-09 2004-04-30 Toshiba Corp Boltzmann model calculation engine for option price, boltzmann model calculation engine for premium, dealing system, dealing program, premium estimation system and premium estimation program
JP2004280309A (en) * 2003-03-13 2004-10-07 Toshiba Corp Boltzmann model calculation engine for reinsurance premium, premium evaluation system and program
US7483840B2 (en) 2002-08-23 2009-01-27 Atera /Solutions Llc Randomized competitive insurance pricing system and method

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004133491A (en) * 2002-08-09 2004-04-30 Toshiba Corp Boltzmann model calculation engine for option price, boltzmann model calculation engine for premium, dealing system, dealing program, premium estimation system and premium estimation program
US7483840B2 (en) 2002-08-23 2009-01-27 Atera /Solutions Llc Randomized competitive insurance pricing system and method
US8612264B2 (en) * 2002-08-23 2013-12-17 Atera Solutions, Llc Randomized competitive insurance pricing system and method
JP2004280309A (en) * 2003-03-13 2004-10-07 Toshiba Corp Boltzmann model calculation engine for reinsurance premium, premium evaluation system and program

Similar Documents

Publication Publication Date Title
Arregui et al. Projecting social contact matrices to different demographic structures
Wilbrecht et al. The influence of setup time on job shop performance
Menascé et al. A methodology for workload characterization of e-commerce sites
Allison et al. Fixed-effects methods for the analysis of nonrepeated events
EP1040413B1 (en) Method for analyzing capacity of parallel processing systems
Seila Introduction to simulation
Campello et al. Queueing models of case managers
Karlstrom et al. A dynamic programming approach to model the retirement behaviour of blue‐collar workers in Sweden
US5841655A (en) Method and system for controlling item exposure in computer based testing
CN115510042A (en) Power system load data filling method and device based on generation countermeasure network
US10313457B2 (en) Collaborative filtering in directed graph
JPH10228462A (en) Path summarizing method by monte carlo simulation
Hogg et al. Gerts qr: A model for multi-resource constrained queueing systems part II: An analysis of parallel-channel, dual-resource constrained queueing systems with homogeneous resources
Keeling et al. Grouping efficiency measures and their impact on factory measures for the machine-part cell formation problem: A simulation study
Gelenbe et al. Random injection control of multiprogramming in virtual memory
US20230177443A1 (en) Systems and methods for automated modeling of processes
Zhang et al. Joint monitoring and prediction of accrual and event times in clinical trials
Fauzan et al. Simulation of agent-based and discrete event for analyzing multi organizational performance
Ali et al. On computing standard errors for marginal structural Cox models
CN111598390B (en) Method, device, equipment and readable storage medium for evaluating high availability of server
CN108804640B (en) Data grouping method, device, storage medium and equipment based on maximized IV
Browne et al. The interaction of multi-programming job scheduling and CPU scheduling
Carroll et al. Alternative estimation procedures for event-history analysis: A monte carlo study
Schechter On the use of computer simulation for research
JP2004334737A (en) Credit risk model deciding device, program, and credit risk model deciding method