Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/169213
Authors: 
Li, Xinjue
Zbonakova, Lenka
Härdle, Wolfgang
Year of Publication: 
2017
Series/Report no.: 
SFB 649 Discussion Paper 2017-023
Abstract: 
In the present paper we propose a new method, the Penalized Adaptive Method (PAM), for a data driven detection of structure changes in sparse linear models. The method is able to allocate the longest homogeneous intervals over the data sample and simultaneously choose the most proper variables with help of penalized regression models. The method is simple yet exible and can be safely applied in high-dimensional cases with di erent sources of parameter changes. Comparing with the adaptive method in linear models, its combination with dimension reduction yields a method which selects proper signi cant variables and detects structure breaks while steadily reduces the forecast error in high-dimensional data. When applying PAM to bond risk premia modelling, the locally selected variables and their estimated coefficient loadings identified the longest stable subsamples over time align with the true structure changes observed throughout the market.
Subjects: 
SCAD penalty
propagation-separation
adaptive window choice
multiplier bootstrap
bond risk premia
JEL: 
C13
C20
E37
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.