Skip to main content

Table 8 Determinants of the Copycat Score

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

 

(1)

(2)

(3)

(4)

(5)

(6)

(7)

IC

  

0.035***

0.065***

0.074***

0.068**

0.054**

Alpha

  

0.419***

0.360***

0.375***

0.380***

 − 0.340**

R2

  

 − 0.055***

 − 0.066***

 − 0.035**

 − 0.032**

0.018

Active

    

0.038***

0.031***

0.036***

Tracking

    

0.027

0.020

0.020

Peers

   

 − 0.055***

 

 − 0.032***

 − 0.040***

REV

0.005

 − 0.004

    

0.002

MOM

0.004

 − 0.003

    

0.010

Flow

0.002

0.002

    

0.003

Exp Ratio

0.713***

0.704***

    

0.177

Turn Ratio

 − 0.002***

 − 0.002***

    

 − 0.005***

logTNA

0.001***

0.001***

    

0.001***

Fund Age

 − 0.0002***

 − 0.0002***

    

 − 0.0001***

PerCom

 − 0.0002***

 − 0.0002***

    

 − 0.0003***

PerCash

0.0002***

0.0002***

    

0.0002

Recession

 

 − 0.008***

    

 − 0.009***

Observations

69,903

69,903

77,516

53,965

22,884

22,249

16,763

Adjusted R2

0.024

0.029

0.057

0.081

0.085

0.092

0.147

  1. The absolute value of the quarterly Copycat measure, obtained from 2000 to 2016, is regressed on measures of fund outperformance or manager skill, and lagged fund characteristics. Measures of skill include 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). Also included are past R2 and fund Alpha obtained from a four-factor model following Amihud and Goyenko (2013), Active Share and Tracking Error (Cremers and Petajisto 2009), and the number of Peers or competing funds (Hoberg et al. 2017) as a percentage of all mutual funds in each period of time. The data on the Copycat score, alpha and R2 is from 2000 to 2016. Data for Active Share and Tracking Error ends in 2009. Data on buy-side competing peer funds runs through 2012. Fund characteristics include measures of return reversal (‘REV’) and momentum (‘MOM’), which consist of the fund’s return in quarter t-1, and the accumulated return from quarter t-4 to t-2. In addition we use the preceding quarter’s fund flows, Expense Ratio, Turnover Ratio, the log of Total Net Assets (‘logTNA’), the fund’s age, and the percentages of fund assets invested in common equity (‘PerCom’) and cash (‘PerCash’). Some models include a dummy variable that takes the value of 1 if the time period lies within an NBER-identified recession. The data for these variables spans 2000–2016. Standard errors are double clustered at the fund and quarter level, and a HAC covariance matrix is used. Statistical significance is denoted by ***, ** and * for significance at the 1%, 5% and 10% levels, respectively