Preliminary communication
Long memory in the Croatian and Hungarian stock market returns
Mejra Festić
; Bank of Slovenia, Ljubljana, Slovenija
Alenka Kavkler
; Department of Quantitative Economic Analysis, Faculty of Economics and Business, University of Maribor, Maribor, Slovenia
Silvo Dajčman
; Department of Finance, Faculty of Economics and Business, University of Maribor, Maribor, Slovenia
Abstract
The objective of this paper is to analyze and compare the fractal structure of the Croatian and Hungarian stock market returns. The presence of long memory components in asset returns provides evidence against the weak-form of stock market effi ciency. The starting working hypothesis that there is no long memory in the Croatian and Hungarian stock market returns is tested by applying the Kwiatkowski-Phillips-Schmidt-Shin (KPSS) (1992) test, Lo’s (1991) modified rescaled range (R/S) test, and the wavelet ordinary least squares (WOLS) estimator of Jensen (1999). The research showed that the WOLS estimator may lead to different conclusions regarding long memory presence in the stock returns from the KPSS and unit root tests or Lo’s R/S test. Furthermore, it proved that the fractal structure of individual stock returns may be masked in aggregated stock market returns (i.e. in returns of stock index). The main fi nding of the paper is that both the Croatian stock index Crobex and individual stocks in this index exhibit long memory. Long memory is identifi ed for some stocks in the Hungarian stock market as well, but not for the stock market index BUX. Based on the results of the long memory tests, it can be concluded that while the Hungarian stock market is weakform efficient, the Croatian stock market is not.
Keywords
Stock market; long memory; efficient-market hypothesis
Hrčak ID:
83353
URI
Publication date:
19.6.2012.
Visits: 1.633 *