Title: Parameter regimes in partially functional linear regression for panel data
Abstract: We introduce a novel semiparametric partially functional linear regression model for panel data. The parametric model part is completely time varying, whereas the functional non-parametric component is allowed to vary over a set of different (functional) parameter regimes. These parameter regimes are assumed latent and need to be estimated from the data additionally to the unknown model parameters. We develop asymptotic theory for the suggested estimators including rates of convergence as n, T → ∞. Our statistical model is motivated from economic theory on asset pricing. It allows to identify different risk regimes, governing the pricing of idiosyncratic risk in stock markets. For our application we develop necessary theoretical ground and offer a vast empirical study based on high-frequency stock-level data for the S&P 500 Index.
Publication Year: 2017
Publication Date: 2017-01-01
Language: en
Type: book-chapter
Indexed In: ['crossref']
Access and Citation
Cited By Count: 1
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot