Investigation of winner-loser portfolio anomalies and size effect anomalies in LQ45 index, Indonesia stock exchange

Abstract

This study investigates the anomalies of winner-loser portfolios and anomalies of size effect on LQ45 stocks in the Indonesia Stock Exchange (IDX), period 2015-2020. The paired sample t-test (parametric) and Wilcoxon test (non-parametric) are used to test samples. Sample selection is carried out using the non-probability and purposive sampling methods. The shares of companies that met the sample criteria were 26. The primary data used cumulative abnormal returns and market capitalization as proxies for a company's stock size. The anomaly of the winner-loser effect phenomenon is observed by dividing the data into two periods: the formation period and the testing period. The periods are both short-term and long-term. The short-term period is six months, and the long-term is 12 months. The first finding shows reversals in average cumulative abnormal returns (CARs) on the winner-loser stock portfolio. In the testing period, the average CARs were turned down, and many average CARs became negative. Portfolio loser in the testing period, the average CARs reversed to positive. The second is a reversal in average CARs in winner-loser stock portfolios caused by the size effect. The performance of small winner-loser portfolios is proven to outperform big winners' portfolios. The reversal occurs in both the short-term and long-term. This finding suggests that behavioral finance can explain this market anomaly, and it is helpful for investors in coming up with strategies to get abnormal returns.