| Abstract: | 本研究檢定共同基金的績效與淨值波動性之關係在定期定額 (AC) 與單筆投資 (LS) 策略間的差異，推論AC法在淨值波動較大時，比LS法有較高機會可降低平均成本，使績效提高。此外，也再行檢視AC與LS的績效是否有差異，雖然過去文獻對此已多有研究，但通常只比較績效本身，而未控制其他影響績效的因素，故可能得到錯誤的結論。本研究以國內發行的開放型共同基金於2000年到2021年的月資料為樣本，績效指標使用風險調整前和調整後報酬率，前者是買入並持有的期間報酬率；後者包括夏普指標及考慮下方風險的索提諾指標。分析持有12、24和36個月的績效，控制基金的風險收益等級、成本費用率、周轉率、總淨值規模和淨流入的影響後，發現淨值波動對績效的影響因持有月數和績效指標之不同而有差異，報酬是否經風險調整有不同的結果；持有36個月時，AC法比LS法的績效較受淨值波動的影響，尤其是風險調整後的指標，而持有12和24個月時，LS法的績效較受淨值波動影響。此結論不能應用於風險收益等級，即風險收益等級與績效之關係，和淨值波動性與績效之關係並不相同。基金績效之比較，三個績效指標之原始值平均數不論持有月數為何都是LS比AC策略顯著較高。但控制其他因素的影響後，不論何種指標，持有36個月都以AC法高於LS法，甚至調整風險後指標在持有24個月已以AC顯著較高。因此，共同基金的投資若採定期定額法，應較長期的持有，至少二年以上，而三年以上更佳，淨值波動大反而有較佳的降低平均成本效果。若持有不超過二年，則應採用單筆投資法。
This study examines the difference between dollar-cost averaging (AC) and lump-sum investing (LS) strategies for the relationship between the performance of mutual funds and the volatility of net asset value. This research infers that the AC method has a higher chance than the LS method to reduce the average cost and improve the performance when the net value fluctuates greatly. Moreover, this study also examines whether the performance of AC and LS is different. Although there have been many studies on this issue in the past, they usually only compare performance itself without controlling for other factors that affect performance, so erroneous conclusions may be drawn.This study uses monthly data from 2000 to 2021 for open-end mutual funds issued in Taiwan as a sample. The performance indicators are not risk-adjusted and risk-adjusted rates of return. The former is the buy-and-hold period return (BHPR); the latter includes Sharpe index (SHR) and Sortino index (SOR) considering downside risk. After analyzing the performance of holding 12, 24 and 36 months, and controlling the impact of the fund's risk-return level, cost-expense ratio, turnover ratio, total net asset value and net flow, it is found that the impact of net asset fluctuations on performance depends on the number of holding months and the performance indicators, and whether the returns are risk-adjusted has different results. When holding for 36 months, the performance of the AC method is significantly affected by the fluctuation of net asset value than the LS method, especially the risk-adjusted indicators, while the holding 12 and 24 months, the performance of the LS method is more affected by fluctuations in net asset. This conclusion can not be applied to the risk-return level, that is, the relationship between risk-return level and performance is not the same as the relationship between net asset volatility and performance.In comparison of fund performance, the average raw value of BHPR, SHR and SOR is significantly higher for LS than for AC strategy regardless of the number of holding months. However, after controlling for the influence of other factors, no matter what kind of indicators, the AC method is higher than the LS method after holding for 36 months, and even after adjusting the risk, the indicator is significantly higher than the AC method after holding for 24 months. Therefore, if the investment of mutual funds adopts the AC method, they should be held for a longer period of time, at least two years or more, and preferably three years or more. If the net asset value fluctuates greatly, it will have a better effect of reducing the average cost. If the holding is not more than two years, the lump sum investment method should be used. |
