Abstract
A (conservative) test is applied to investigate the optimal lag structure for modelingrealized volatility dynamics. The testing procedure relies on the recent theoretical results that showthe ability of the adaptive least absolute shrinkage and selection operator (adaptive lasso) to combinee cient parameter estimation, variable selection, and valid inference for time series processes. In anapplication to several constituents of the SP (ii) in many cases the relevant information for prediction is included in the first 22lags, corroborating previous results concerning the accuracy and the diffculty of outperforming outof-sample the heterogeneous autoregressive (HAR) model; and (iii) some common features of theoptimal lag structure can be identified across assets belonging to the same market segment or showinga similar beta with respect to the market index.
Highlights
For many years the prediction of financial assets has been the goal of extensive research
While we can see that the sector and the beta have an influence on the lag structure that is selected by the adaptive lasso, we find that other factors such as liquidity and value do not seem to show visible effects
The goal of this paper was to analyze whether the structural assumptions that the heterogeneous autoregressive (HAR) model implies can be recovered using solely statistical techniques
Summary
For many years the prediction of financial assets has been the goal of extensive research. While Craioveanu and Hillebrand (2010) have shown that it is only of little importance what investment horizons are assumed for the different investor groups, that is which aggregation frequencies are chosen in a HAR-like structure, Wang et al (2013) show that an ARFIMA model subject to structural breaks in the mean and the memory parameters of the process can be approximated well by an AR model. The present paper is an extension of the work by Audrino and Knaus (2014) who analyze the validity of the lag structure implied by the HAR assumption from a model selection perspective. The content of this paper can be summarized as follows: Section 2 presents the theoretical foundations of both the HAR model and the adaptive lasso estimator and introduces the hypotheses of interest and the testing procedure. The concept of multiple hypothesis testing will be briefly revisited
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have