The Wiley Finance Series : Handbook of News Analytics in Finance

(Chris Devlin) #1

204 News and abnormal returns


Table 7.6.Percentage return performance for portfolios of stocks purchased minus portfolios
of stocks sold in partitions based on abnormal volume sorts and the previous day’s return sorts.
Trades data are for investors at a large discount brokerage (LDB—January 1991 through
November 1996), investors at a large retail brokerage (LRB—January 1997 through June
1999), and investors at a small discount brokerage (SDB—January 1996 through June 15,
1999). For investors at each brokerage, we form two portfolios for each partition: stocks purchased
and stocks sold. Stocks enter the portfolio the day following the purchase or sale. Portfolios are re-
balanced daily assuming a holding period of 21 trading days (i.e., 1 month). The purchase and sale
portfolios are constructed using the value of purchases and sales, respectively. We evaluate the
difference in the return of these two portfolios (RbtRst). We estimate the following monthly time-
series regression:


Rbt¼ (^) jþ (^) jðRmtRftÞþsjSMBtþhjVMGtþmjWMLtþ"jt;
whereRftis the monthly return on T-Bills,Rmtis the monthly return on a value-weighted
market index,SMBtis the return on a value-weighted portfolio of small stocks minus the
return on a value-weighted portfolio of big stocks,VMGtis the return on a value-weighted
portfolio of high-book-to-market (value) stocks minus the return on a value-weighted portfolio
of low-book-to-market (growth) stocks, andWMLtis the return on a value-weighted portfolio
of recent winners minus the return on a value-weighted portfolio of recent losers. The same re-
gression is estimated for portfolioRstand for the difference in these two portfolios (RbtRst).
Panels A, B, and C report four-factor alpha portfolios based on all trades at each brokerage
and for the combined sample sorted on measures of attention. In panel A, stocks first are
sorted daily on the basis of the current day’s abnormal trading volume prior to forming pur-
chase and sale portfolios. Abnormal trading volume is calculated as the ratio of the current
day’s trading volume (as reported in the CRSP daily stock return files for NYSE, ASE, and
NASDAQ stocks) divided by the average trading volume over the previous 252 trading days.
In panel B, stocks are first sorted daily into deciles on the basis on the previous day’s return
as reported in the CRSP daily stock return files for NYSE, ASE, and NASDAQ stocks. In
panel C, stocks first are sorted daily on the basis of the current day’s abnormal trading
volume and then again sorted daily into deciles on the basis on the previous day’s return.
Deciles 1–2, 3–8, and 9–10 are combined for each sorting criteria prior to forming purchase
and sale portfolios.
Panel A: Percentage four-factor alphas for purchase, sales, and purchase less sales portfolios
formed after sorting on the current day’s abnormal trading volume—combined (2/91 to 6/99)
Abnormal volume Buys Sells Buys–sells
sort decile
Alpha t-statistic Alpha t-statistic Alpha t-statistic
1 (low) 0.008 0.010 0.037 0.110 0.029 0.060
2 0.836 3.250 0.800 4.130 0.036 0.150
3 0.162 0.710 0.262 1.530 0.100 0.570
4 0.134 0.680 0.186 1.090 0.319 2.370
5 0.095 0.460 0.069 0.380 0.026 0.210
6 0.319 1.650 0.256 1.510 0.064 0.610
7 0.364 1.880 0.288 1.480 0.075 0.760
8 0.557 2.480 0.514 2.440 0.043 0.390
9 0.286 1.100 0.480 1.850 0.195 1.450
10 (high) 0.739 2.290 0.049 0.150 0.690 3.830

Free download pdf