News & Events


Events Detail

​Economics Seminar | Estimation and Inference of Fractional Continuous-time Model with Discrete-Sampled Data 

Published on: 20-Sep-2019

EventEstimation and Inference of Fractional Continous-time Model with Discrete-Sampled Data
Prof Xiaohu WANG
Assistant Professor, Department of Economics,
The Chinese University of Hong Kong
Date20 Sep 2019 (Fri)
Time3:30pm – 5:00pm
VenueHSS Meeting Room 5 (HSS-04-94)

About the Seminar
This paper proposes a two-stage method for estimating parameters in a parametic fractional continous-time model based on discrete-sampled observations. In the first stage, the Hurst parameter is estimated based on the ratio of two second-order differences of observations taking at different time scales. In the second stage, the other parameters are estimated by method-of-moments. All estimators have closed-form expressions and are easy to obtain. Large sample theory of the proposed estimators is derived under either the in-fill asymptotic scheme or the double asymptotic scheme. Extensive simulation studies show that the proposed theory performs well in finite samples. Two empirical studies are carried out. The first empirical study, based on the daily realized volatility is too rough for continous-time models driven by a standard Brownian motion. The econd empirical study is for the daily realized volatility of exchange rates over 1986-1999. The estimate of the Hurst parameter is again much less than one half. Moreover, it is found that the proposed fractional continous-time model performs better than the ARFIMA model out-of-sample. 

Back to listing