Fractional Cointegration of Daily High and Low Stock Prices
|Author:||Mgr. Dvořáková Sylvie|
|Year:||2014 - summer|
|Leaders:|| doc. PhDr. Jozef Baruník Ph.D.
|Work type:|| Doctoral
|Awards and prizes:|
|Abstract:||In this thesis we provide unique empirical support for the fractional cointegration of daily high and low stock prices. The recently formalized fractionally cointegrated VAR model
by Johansen and Nielsen (2012) is used due to its ability to
capture both the cointegration between daily high and low stock prices and the long memory of their linear combination,
the range. Daily high and low stock prices are of particular interest as they provide valuable information about range
- based volatility, which is considered a highly efficient and robust estimator of volatility. We provide a comparison of the
Czech PX index with the developed market indices (DAX, FTSE 100, S&P 500 and NIKKEI225) during the 2003 - 2012 period
as well as before and after the crisis. We find that the range of all indices displays long memory and is mostly in the non - stationary region (except for the ranges of the PX Index and NIKKEI 225 indices in the pre - crisis period). These findings provide evidence thatvolatility may not be a stationary process. No common pattern is detectedamong all five market indices and different behaviour is also observed in the pre-crisis and post-crisis periods.