Specification and Structural Break Tests for Additive Models with Applications to Realized Variance Data

Posted: 1 Nov 2013 Last revised: 20 Jun 2015

See all articles by Matthias R. Fengler

Matthias R. Fengler

University of St. Gallen - School of Economics and Political Science; Swiss Finance Institute

Enno Mammen

University of Mannheim - Department of Economics

Michael Vogt

University of Cambridge

Date Written: November 1, 2013

Abstract

We study two types of testing problems in a nonparametric additive model setting: We develop methods to test (i) whether an additive component function has a given parametric form and (ii) whether an additive component has a structural break. We apply the theory to a nonparametric extension of the linear heterogeneous autoregressive model which is widely employed to describe realized variance data. We find that the linearity assumption is often rejected, but actual deviations from linearity are mild.

Keywords: Additive models, Backfitting, Nonparametric time series analysis, Specification tests, Realized variance; Heterogeneous autoregressive model

JEL Classification: C14, C58

Suggested Citation

Fengler, Matthias R. and Mammen, Enno and Vogt, Michael, Specification and Structural Break Tests for Additive Models with Applications to Realized Variance Data (November 1, 2013). Journal of Econometrics, Vol. 188, No. 1, 2015, pp. 196-218, Available at SSRN: https://ssrn.com/abstract=2348600 or http://dx.doi.org/10.2139/ssrn.2348600

Matthias R. Fengler (Contact Author)

University of St. Gallen - School of Economics and Political Science ( email )

Bodanstrasse 6
CH-9000 St. Gallen, 9000
Switzerland

HOME PAGE: http://www.mathstat.unisg.ch/fengler

Swiss Finance Institute ( email )

c/o University of Geneva
40, Bd du Pont-d'Arve
CH-1211 Geneva 4
Switzerland

Enno Mammen

University of Mannheim - Department of Economics ( email )

Mannheim, 68131
Germany

Michael Vogt

University of Cambridge ( email )

Trinity Ln
Cambridge, CB2 1TN
United Kingdom

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

Paper statistics

Abstract Views
648
PlumX Metrics