Skip to main content

Table 12 Performance implications

From: Does communication increase investors’ trading frequency? Evidence from a Chinese social trading platform

  \(Return_{i,t}\)
  (1) (2) (3) (4)
\(Leader\ comment_{i,t-1}\) − 0.0007\(^{***}\) − 0.0007\(^{***}\)   
(− 2.80) (− 2.73)   
\(Leader\ count_{i,t-1}\)    − 0.00004 − 0.00003
   (− 0.27) (− 0.22)
\(Leader\ positive_{i,t-1}\)    − 0.0022 − 0.0021
   (− 0.58) (− 0.55)
\(Leader\ negative_{i,t-1}\)    − 0.0030 − 0.0027
   (− 0.86) (− 0.77)
\(Trades_{i,t-1}\) − 0.0023\(^{***}\)   − 0.0021\(^{***}\)  
(− 8.78)   (− 7.70)  
\(Turnover_{i,t-1}\)   −0.0025\(^{***}\)   −0.0025\(^{***}\)
  (− 10.57)   (− 9.35)
\(Return\ SD_{i,t-1}\) 0.0025 0.0044 − 0.0007 0.0010
(0.34) (0.59) (− 0.07) (0.10)
\(No.securities_{i,t-1}\) 0.0006 0.0003 0.0008 0.0006
(0.75) (0.45) (0.97) (0.72)
\(No.followers_{i,t-1}\) −0.0058\(^{***}\) − 0.0057\(^{***}\) − 0.0050\(^{***}\) − 0.0049\(^{***}\)
(− 7.02) (− 7.01) (− 6.09) (− 6.12)
\(Portfolio\ age_{i,t-1}\) 0.0004 0.0004 0.0018\(^{***}\) 0.0017\(^{***}\)
(1.19) (1.21) (3.30) (3.20)
\(No.leaders_{i,t-1}\) 0.0001 0.0001 − 0.0002 − 0.0002
(0.27) (0.18) (− 0.43) (− 0.51)
\(Leader\ return_{i,t-1}\) − 0.0034 − 0.0031 − 0.0078 − 0.0074
(− 0.58) (− 0.53) (− 1.03) (− 0.97)
\(Leader\ SD_{i,t-1}\) − 0.0070 − 0.0063 − 0.0212 − 0.0209
(− 0.56) (− 0.51) (− 1.45) (− 1.43)
\(Leader\ trades_{i,t-1}\) 0.0003 0.0003 0.0002 0.0002
(1.07) (1.10) (0.63) (0.66)
\(\text{Leader followers}_{i,t-1}\) − 0.000002 − 0.00002 − 0.0001 − 0.0002
(− 0.01) (− 0.08) (− 0.56) (− 0.58)
\(Leader\ securities_{i,t-1}\) − 0.0003 − 0.0003 − 0.0004 − 0.0003
(− 1.10) (− 1.11) (− 0.93) (− 0.82)
\(Leader\ age_{i,t-1}\) 0.0001 0.0001 0.0001 0.0001
(0.46) (0.51) (0.62) (0.56)
Portfolio fixed effects Yes Yes Yes Yes
Time fixed effects Yes Yes Yes Yes
Observations 262,457 262,457 150,447 150,447
Adjusted \(R^2\) 0.2043 0.2062 0.2314 0.2332
  1. This table presents the estimation results of fixed-effects panel regressions on portfolio returns. Only treated real-account portfolios are included in the regressions. All explanatory variables are lagged by one week. Standard errors are double-clustered at the portfolio level and over time. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively