Testing for non-linearity in multivariate stochastic processes

Two well known multivariate non-linearity tests are modified using a principal component analysis. The Monte Carlo results show that the proposed principal component-based tests do provide a remarkable dimensionality reduction without any systematic power loss. It can be concluded that using linear dynamic economic models is in sharp contrast with our empirical findings.

MoreLess

Year of publication:	2013-09
Authors:	Vavra, Marian
Institutions:	Národná Banka Slovenska
Subject:	non-linearity testing \| principal component analysis \| Monte Carlo method

More details

Extent:	application/pdf
Series:	Working and Discussion Papers. - ISSN 1337-5830.
Type of publication:	Book / Working Paper
Notes:	Number WP 2/2013 37 pages
Classification:	C12 - Hypothesis Testing ; C15 - Statistical Simulation Methods; Monte Carlo Methods ; C32 - Time-Series Models
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10010734234