Skip to main content

Table 9 Tone of leaders’ comments and followers’ trading frequency - Alternative dictionary

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

 

\(Trades_{i,t}\)

\(Turnover_{i,t}\)

 

(1)

(2)

(3)

(4)

 

FE

SYS-GMM

FE

SYS-GMM

\(Leader\ count_{i,t-1}\)

0.0197\(^{***}\)

0.0221\(^{**}\)

0.0186\(^{***}\)

0.0351\(^{**}\)

(4.35)

(2.13)

(3.24)

(2.48)

\(\text{Leader positive}_{i,t-1}\)

0.0386

− 2.0590

0.0924

− 0.7023

(0.51)

(− 1.58)

(0.82)

(− 0.54)

\(Leader\ negative_{i,t-1}\)

− 0.1029

− 0.8477

− 0.0395

0.3755

(− 1.26)

(−0.76)

(− 0.34)

(0.27)

\(Return_{i,t-1}\)

0.6394\(^{***}\)

0.6918\(^{***}\)

0.0698

0.7521\(^{***}\)

(5.86)

(4.24)

(0.47)

(3.86)

\(Return\ SD_{i,t-1}\)

0.4159\(^{**}\)

2.0980\(^{**}\)

0.8850\(^{***}\)

1.9590

(2.07)

(2.10)

(2.95)

(1.07)

\(No.securities_{i,t-1}\)

0.4640\(^{***}\)

0.0092

0.2987\(^{***}\)

− 0.0429\(^{**}\)

(23.84)

(0.86)

(15.03)

(− 2.55)

\(No.followers_{i,t-1}\)

0.0694\(^{**}\)

0.7378\(^{**}\)

0.0845\(^{***}\)

0.0730\(^{*}\)

(2.09)

(2.44)

(2.80)

(1.91)

\(Portfolio\ age_{i,t-1}\)

− 0.1435\(^{***}\)

−0.0213\(^{***}\)

−0.1412\(^{***}\)

−0.0200\(^{*}\)

(− 6.54)

(−2.92)

(−4.77)

(− 1.73)

\(No.leaders_{i,t-1}\)

0.0921\(^{***}\)

0.0225

0.0708\(^{***}\)

− 0.0096

(4.50)

(1.20)

(3.50)

(− 0.43)

\(Leader\ return_{i,t-1}\)

0.2166\(^{*}\)

0.0225\(^{**}\)

0.3518\(^{**}\)

1.0923

(1.92)

(2.12)

(2.56)

(0.57)

\(Leader\ SD_{i,t-1}\)

0.5769

1.9864

0.4285

5.3585\(^{**}\)

(1.55)

(1.15)

(1.05)

(2.25)

\(Leader\ trades_{i,t-1}\)

0.0504\(^{***}\)

− 0.0290

0.0365\(^{**}\)

0.0158

(4.43)

(− 1.02)

(2.26)

(0.38)

\(Leader\ followers_{i,t-1}\)

0.0132

0.0049

0.0117

− 0.0115

(1.25)

(0.54)

(0.87)

(− 0.89)

\(Leader\ securities_{i,t-1}\)

0.0172

− 0.0010

0.0364

− 0.0130

(1.06)

(− 0.08)

(1.53)

(− 0.62)

\(Leader\ age_{i,t-1}\)

− 0.0380\(^{***}\)

− 0.0182

− 0.0416\(^{***}\)

− 0.0152

(−3.47)

(− 1.49)

(− 2.94)

(− 0.80)

\(Y_{i,t-1}\)

 

0.8229\(^{***}\)

 

0.6755\(^{***}\)

 

(12.40)

 

(5.51)

\(Y_{i,t-2}\)

 

− 0.0780

 

− 0.0514

 

(− 0.50)

 

(− 0.33)

\(Y_{i,t-3}\)

 

0.0917

 

0.0533

 

(0.91)

 

(0.42)

\(Y_{i,t-4}\)

 

− 0.0078

 

0.1039

 

(− 0.63)

 

(0.72)

\(Y_{i,t-5}\)

 

− 0.0058

 

− 0.0091

 

(− 0.82)

 

(− 0.34)

Portfolio fixed effects

Yes

Yes

Yes

Yes

Time fixed effects

Yes

Yes

Yes

Yes

Observations

150,447

140,419

150,447

140,419

Adjusted \(R^2\)

0.3521

0.3566

AR(1) test (p value)

0.000

0.000

AR(2) test (p value)

0.158

0.427

Hansen test of over-identification (p value)

0.449

0.446

Diff-in-Hansen test of exogeneity (p value)

0.539

0.598

  1. This table reports the results from the fixed-effects (FE) estimation of the panel regression model specified in Eq. 2 (odd columns) and the GMM (SYS-GMM) estimation of the panel regression model specified in Eq. 3 (even columns). The dependent variable is either the (log) number of trades of portfolios (Columns 1 and 2) or the turnover ratio of portfolios (Columns 3 and 4).
  2. \(Leader\ Positive\) and \(Leader\ Negative\) are constructed based on NTUSD. Only treated real-account portfolios are included in the regressions. All explanatory variables are lagged by one week. In odd columns, standard errors estimated by the fixed-effects approach are double-clustered at the portfolio level and over time. In even columns, standard errors estimated by the system GMM approach are clustered at the portfolio level. *, **, and *** denote significance at the 10%, 5%, and 1% level, respectively