INDONESIAN FIXED INCOME MUTUAL FUNDS PERFORMANCE PERSISTENCE (2011-2016)

• Main Author: Anglani (NIM 19014091), Gracia
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/22304
• Institution: Institut Teknologi Bandung

This research examines the performance persistence of fixed income mutual funds in Indonesia. Mutual fund has been developing in Indonesia. Recently, fixed income mutual funds performance has outperformed equity mutual funds, which would be a good signal for investors. A problem arises when commonly investors were still affected by past performance in deciding which mutual fund to invest. On the other side, there is a disclaimer in every mutual funds that past performance does not guarantee future performance, which in line with <br /> <br /> the Efficient Market Hypothesis (EMH). Therefore, this study conducted to investigate it and give suggestions to the investors whether they can rely on past performance. A sample of 68 mutual funds in 2011-2016 collected from OJK, was analyzed by using a contingency table. Malkiel’s Z-test, Brown’s and Goetzmann’s odds ratio (OR), and Kahn’s and Rudd’s X 2-statistic with Yates’ correction were applied to test the robustness of the performance persistence. It was discovered that there was an evidence of performance persistence in Indonesian fixed income mutual funds during the observed periods. In the comparison between a certain period’s performance and its subsequent period performance(s), the majority of the individualobservations have found an evidence of performance persistence with various significance levels. In the other word, some of the test results are significant at 1%, 5%, or 10% significance level, while the others are not. However, if the individual observations were to be merged as a total data, the results are 1% statistical significant. In addition, this research also does a comparison between a certain period’s performance and its historical performance.The findings have found that the usage of 2 years historical data shows a higher significance level of performance persistence compared to the usage of 1 year historical data. The usage of historical data more than 2 years (3 years) does not give significantdifference.