RUS  ENG
Full version
JOURNALS // Computer Research and Modeling // Archive

Computer Research and Modeling, 2012 Volume 4, Issue 3, Pages 631–638 (Mi crm516)

MODELS OF ECONOMIC AND SOCIAL SYSTEMS

Time-clusterring of stock indicies’ big fall

A. M. Kazaryana, A. B. Shapovalb

a Financial University under the Government of the Russian Federation, 49 Leningradsky Prospekt, Moscow, 125993, Russia
b Institute of Earthquake Prediction Theory and Mathematical Geophysics, 84/32 Profsoyuznaya str., Moscow, 117997, Russia

Abstract: The paper estimates the recurrence rate of stock indicies S&P100, CAC40, DAX, FTSE, AMEX, ATX, NASDAQ, BEL20. The introduced qunatitative measure of the recurrence rate underlies type I and type II errors. We show that more than three quarters of the indicies’ falls occur on average during the first quarter of the time between them. This result expands from sufficiently large falls, which are observed on average two times a year, over smaller falls, which occur approximately once 1.5–2 months.

Keywords: inter-event distribution, errors of the first and second kind.

UDC: 51.77

Received: 25.05.2012

DOI: 10.20537/2076-7633-2012-4-3-631-638



© Steklov Math. Inst. of RAS, 2024