Event Prediction Technique Based on Correlation Mapping and Fourier Decomposition

8 Pages Posted: 3 Mar 2023

See all articles by Dmitry Lande

Dmitry Lande

National Technical University of Ukraine - Igor Sikorsky Kyiv Polytechnic Institute

V. Shchutsky

Index Systems ltd.

Ellina Shnurko-Tabakova

Index Systems ltd.

Leonard Strashnoy

University of California, Los Angeles (UCLA)

Date Written: February 27, 2023

Abstract

The method of predicting events based on correlation mapping and Fourier decomposition of relevant time series, which is presented in this paper, is intended for the predictive calculation of signs that an event will occur. As input data for calculating the forecast, in addition to the series that needs to be extended with forecast values, relevant series that are close in the content are used. The calculation is based on a model that calculates the correlations of the total values of the Fourier decomposition for all relevant time series with the historical series. In this case, a certain number of constant parameters of the Fourier decomposition is calculated as the result of solving the optimization problem. Based on the generated model and the values of constant parameters, further values of the initial series are calculated, that is, the predicted event is determined.

Keywords: Event Forecasting, Relevant Time Series, Fourier Decomposition, Correlation of Time Series, Correlation Mapping

JEL Classification: A13, C00, C6, C80

Suggested Citation

Lande, Dmitry and Shchutsky, V. and Shnurko-Tabakova, Ellina and Strashnoy, Leonard, Event Prediction Technique Based on Correlation Mapping and Fourier Decomposition (February 27, 2023). Available at SSRN: https://ssrn.com/abstract=4372251 or http://dx.doi.org/10.2139/ssrn.4372251

Dmitry Lande (Contact Author)

National Technical University of Ukraine - Igor Sikorsky Kyiv Polytechnic Institute ( email )

V. Shchutsky

Index Systems ltd.

Ellina Shnurko-Tabakova

Index Systems ltd.

Leonard Strashnoy

University of California, Los Angeles (UCLA) ( email )

405 Hilgard Avenue
Box 951361
Los Angeles, CA 90095
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
35
Abstract Views
161
PlumX Metrics