‘Smart’ copycat mutual funds: on the performance of partial imitation strategies

Stein, Roberto

doi:10.1186/s40854-022-00398-7

Financial Innovation

Table 4 Mutual fund Copycat Score and future performance

From: ‘Smart’ copycat mutual funds: on the performance of partial imitation strategies

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
Panel A: 1 Year ahead gross return
absCopy Active R2 Alpha Tracking IC Peers	22.922***	7.710***	− 6.487*	76.215*	1.588	4.466*	− 12.525***	15.350* 9.418* − 4.946 116.829* − 16.427 − 15.492** − 6.404*
PerCom	− 0.001	− 0.017	0.000	− 0.009	− 0.027	− 0.009	− 0.028	− 0.008
PerCash	− 0.016	− 0.076***	− 0.023	− 0.026	− 0.047***	− 0.026	− 0.008	− 0.052**
logTNA	− 0.178*	− 0.085	− 0.086	− 0.122	− 0.115	− 0.161	− 0.170	− 0.013
Fund Age	0.045	− 0.158	− 0.052	− 0.037	− 0.286*	− 0.021	− 0.009	− 0.035
Turn Ratio	− 0.513*	− 0.526	− 0.578*	− 0.413	− 0.470	− 0.553**	− 0.873**	− 0.401
Exp Ratio	− 103.016***	− 145.326***	− 112.309***	− 79.032**	− 41.121	− 97.829***	− 104.711***	− 101.53***
Obs	221,226	136,492	210,891	210,891	136,492	219,945	147,031	100,130
Adj. R²	0.63	0.68	0.64	0.63	0.67	0.63	0.69	0.74
Panel B: 1 year ahead alpha
absCopy Active R2	0.574***	0.136**	− 0.380*					0.445* 0.187** − 0.287
Alpha				5.851*				13.198***
Tracking IC					− 0.315	0.550***		− 1.110** − 0.380
Peers							− 0.225	− 0.094
PerCom	− 0.001	0.000	0.001	0.000	0.000	0.000	0.001	0.001
PerCash	0.001	0.000	0.002	0.001	0.001	0.001	0.002**	0.002
logTNA	− 0.004	0.000	− 0.003	− 0.004	0.000	− 0.005	− 0.005	− 0.003
Fund Age	0.016*	0.004	0.017**	0.018**	0.001	0.014*	0.007	0.008
Turn Ratio	− 0.042***	− 0.051***	− 0.044***	− 0.039***	− 0.047***	− 0.042***	− 0.059***	− 0.054***
Exp Ratio	− 10.493***	− 11.125***	− 11.613***	− 9.389***	− 8.532***	− 11.874***	− 11.448***	− 9.660***
Obs	221,226	136,492	210,891	210,891	136,492	219,945	147,031	100,130
Adj. R²	0.10	0.13	0.14	0.12	0.13	0.12	0.14	0.20

Mutual fund performance measures are regressed on lagged fund characteristics and measures previously identified as predictors of future fund performance or fund manager skill. In Panel A the dependent variable is the fund’s annual gross return calculated using monthly data from month t + 1 to t + 12, while in Panel B it’s the fund’s alpha obtained from a Carhart (1997) four-factor model estimated using the same data span. Regressors are obtained using data available at the end of month t or before. ‘absCopy’ is the absolute value of the fund’s Copycat Score. IC is the fund’s industry concentration, estimated as the standard deviation of the percentages of assets allocated by each fund to 10 industry groups, as tabulated in Kacperczyk et al. (2005). Past R² and fund Alpha are obtained from a four-factor model following Amihud and Goyenko (2013). Active Share and Tracking Error are used as in Cremers and Petajisto (2009). The number of Peers or competing funds (Hoberg et al. 2017) is included as a percentage of all mutual funds in each period of time. Control variables include fund characteristics, such as the percentages of fund assets invested in common equity (PerCom) and cash (PerCash), the log of Total Net Assets (‘logTNA’), the fund’s age, Turnover Ratio, and Expense Ratio. For most models and variables the data spans 2000–2016. Data for Active Share and Tracking Error ends in 2009. Data on Peers runs through 2012. The Fama-MacBeth regression methodology is used, with Newey-West standard errors. Statistical significance is denoted by ∗∗∗, ∗∗ and ∗ for significance at the 1%, 5% and 10% levels, respectively.

Back to article page