Research On Persistence In Market Timing Ability Of Bond Fund Managers

Related research on mutual funds has been conducted to provide the certaintrading strategies with theoretical foundation and empirical evidence, as well as animportant reference for the supervisors. In this paper, the author decompose theexcess return of the bond mutual fund into market timing ability and individual bondselection ability, and then focuses on the persistence in market timing ability of bondfund managers. The ways used to measure the market timing ability of bond fundmanagers include: T-M Model/H-M Model, Asset Allocation Method and DurationMethod, for the short-term persistence, mid-term persistence and long-termpersistence respectively. To see whether the persistence exists or not, the author runsregression model and examines if the coefficient of X is significantly positive.The result is that while the short-term persistence exists and the mid-termpersistence may exist, the long-term persistence does not exist. This result has at leasttwo meanings:1) Investors may exploit excess return by selecting some bond fundswith good recent historical performance and constructing their investment portfolio,which is called a winner-picking strategy;2) While the market efficiency hypothesisdoes not hold for short-term or mid-term, it holds for long-term.After getting the result, the author conducts future research to see if thewinner-picking strategy is profitable. It is found that the profitability of thewinner-picking strategy is influenced by the referring period, the selection criteria, theminimum of the scale of the fund picked out and the number of the funds included inthe investment portfolio. The general conclusion here is that investors could usewinner-picking strategy to exploit excess returns if it is implemented properly due tothe existence of short-term persistence in market timing ability of bond fundmanagers.