Abstract
High frequency risk measures: An empirical study of the Indian stock market
by
Saranjeet Kaur Bhogal
The purpose of this paper is to compute intraday high-frequency risk (HFR) measures for market risk of some of the selected stocks listed on the National Stock Exchange of India (NSE). We compute value-at-risk (VaR) accompanied with a time-at-risk (TaR) measure. An autoregressive conditional duration (ACD) model is applied to model the duration dynamics whereas, the volatility is captured using generalized autoregressive conditional heterocedastic (GARCH) models. The validity of the VaR and TaR forecasts is assessed using a backtesting procedure. The stocks are selected from three different sectors: IT-software sector, cement and construction materials sector and pharmaceuticals and drugs sector. Selected stocks from the pharmaceuticals and drugs sector show frequent small gains and few extreme losses. Whereas selected stocks from IT-software sector (except one) and cement and construction materials sector (except one) show frequent small losses and few extreme gains. Comparable results are obtained when median (instead of arithmetic mean) is used for aggregation of transactions having the same time stamp for one of the stock.
Keywords. High-frequency risk measure, National Stock Exchange of India.
Committee
Workshop
Key Dates
Communication
StatFin Main Webpage