Analysis of efficent markets hypothesis in share market with long memory models

The COVID-19 pandemic has precipitated significant alterations not only in global lifestyles and daily routines but also in investment behaviors. This study aims to test the validity of the efficient markets hypothesis in the indices in Borsa Istanbul during the COVID-19 pandemic and the period when the pandemic effects on the country's economies and individuals continued. The study involves analyzing the stationarity of the return series for these equity indexes using both traditional unit root tests (ADF and PP) and those considering structural breaks (Zivot Andrews). Furthermore, the presence of the EMH was examined through the application of long memory models, employing methods such as Geweke and Porter-Hudak, Modified Log-Periodogram, and Robinson Gaussian Semi-Parametric. The findings reveal that the returns of BIST100, BIST50, BIST30, and BISTTUM indexes are non-stationary, do not exhibit mean reversion following shocks, and lack long memory characteristics. These results suggest that investors' attempts to predict price movements of these indexes may be misleading, indicating that only weak-form efficiency could be argued for these indexes.