JP2009266184A

JP2009266184A - Markov risk calculation

Info

Publication number: JP2009266184A
Application number: JP2008136035A
Authority: JP
Inventors: Izumi Yamazaki; 泉山▲崎▼; Junichi Yamazaki; 淳一山▲崎▼
Original assignee: Individual
Current assignee: Individual
Priority date: 2008-04-23
Filing date: 2008-04-23
Publication date: 2009-11-12

Abstract

<P>PROBLEM TO BE SOLVED: To obtain an effective calculation result within an appropriate amount of time when calculating the risk of an interest by obtaining the transition route of interest transition by a transition probability matrix in terms of a probability and a brute force. <P>SOLUTION: Each transition route is calculated by grouping using vectors in the order of a larger interest or a smaller interest indicated by a state included by each transition route, and each transition route is ordered by successively being changed from a state that an interest load is large to a state that it is small or from a state that the interest load is small to a state that it is large from the back of the vectors, and calculation is advanced according to the order. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

本発明は金利予測のリスク計算を推移確率マトリクスを用いてで総当り的に行うとき、膨大な計算量となって妥当な時間内に有効な計算結果を得ることが困難であることを解消するためのものであるThe present invention eliminates the difficulty of obtaining an effective calculation result within a reasonable amount of time when the risk calculation of interest rate prediction is performed brute-force using a transition probability matrix. Is for

従来は推移確率マトリクスで金利推移の推移ルートを確率的に且つ総当り的に求めて金利のリスク計算を行うと、全ルートについて計算を行わないと結果が得られないので計算時間がかかりすぎ、使用可能な結果を得ることが困難であった。Conventionally, if you calculate the interest rate transition route stochastically and brute-force with the transition probability matrix and calculate the risk of interest rate, the result will not be obtained unless you calculate for all routes, so it takes too much time, It was difficult to obtain usable results.

従来は推移確率マトリクスで金利推移の推移ルートを確率的に且つ総当り的に求めて金利のリスク計算を行うことは計算時間がかかりすぎ、使用可能な結果を得ることがて困難であった。Conventionally, it has been difficult to obtain a usable result because it takes too much calculation time to calculate the risk of interest rate by probabilistically and omnidirectionally determining the transition route of interest rate transition using the transition probability matrix.

各推移ルートをおのおのが含む状態が示す金利の大きいもの順にまたは小さいもの順に並べたベクトルによりグループ化して計算を行い、ベクトルの後ろから順次金利負担の大きい状態から小さいもの、又は小さいものから大きいものに変更することによって順序付けし、その順序に従って計算を進めていくこと。Grouped by vectors arranged in the order of increasing or decreasing interest rates indicated by the states that each transition route includes, and calculating from the back of the vector, the interest rate burden from the largest to the smallest, or from the smallest to the largest Order by changing to, and proceed with the calculation according to the order.

従来は推移確率マトリクスで推移ルートについて総当り的に金利負担額と確率を計算すると、妥当な時間内に全計算をやりつくすことが出来ず、リスクの把握計算として使用できなかったが、本発明を使用すると、与えられた時間内に一定の金利負担額を超える又は下回る確率はなんパーセントという具合に、使用していい時間に応じて使用可能な数値が得られるので、時間が多ければ高い精度で、少なければ少ない精度で、ともかく使用可能な結果が得られることとなる。Conventionally, if the interest rate burden and probability are calculated brute-force for the transition route with the transition probability matrix, all calculations could not be exchanged within a reasonable time and could not be used as a risk comprehension calculation. If you use, you can obtain a usable number according to the time you can use, such as what percentage the probability of exceeding or falling below a certain amount of interest burden in a given time, so if there is more time, higher accuracy If it is small, a usable result can be obtained with less accuracy.

本発明で行う計算のフローを例示する。The flow of the calculation performed by this invention is illustrated.

Claims

Considering that the future interest rate trend will change according to the Markov transition probability matrix by the transition probability on the computer, the probability that the interest rate burden within a certain period will be more than or less than a certain amount, while determining the probability and interest rate burden for every transition route In the calculation process to be performed, each transition route is grouped by vectors arranged in descending order of the interest rates indicated by the states included, and calculation is performed from the rear of the vector in order from the largest to the smallest, or from the smallest to the largest. Order by changing and proceed with the calculation according to the order.