5. Regression Analysis. To test whether the immediate market response to earnings surprises announced on a Friday is slower than for those announced on other weekdays, produce a table similar to DellaVigna and Pollet, Table II, Panel A. Use three different specifications for the regressions: (1) CAR[-1/+1) Bo+Frln+ SurQx+BqFlexSurQck+ Bek CAR[-1/+1]= Bo+B,Frk+ SurQ11k+ FrlxSurQ11{k+ Bk (2) CAR(-1/+1] = Bo+BFrlx+SurQTop+k+Frl,zSurQToPER+ Bex (3) In the above regression equations, Fri is a dummy indicating a Friday earnings announcement, SurQ is the bin number of the earnings surprise, i.e. a number from 1 to 11, SurQ11 is a dummy for bin #11 of earnings surprises, and SurQTop is a dummy for the two highest bins of earnings surprises, Q10 and Q11. The indices t and k stand for the fiscal quarter and the firm, respectively. For regression (2) you need to drop all date in bins 2-10 before running the regression. (Why?) For regression (3) you need to drop bins 3-9. For each regression model, run two regressions, one without additional controls and one with what DellaVigna and Pollet call "standard controls (interacted)". If you do not have all the controls in your dataset, just use those that you can construct Table II Stock Price Response to Earnings Announcements for the Top and Bottom Groups Stocks in CRSP are matched to quarterly earnings announcements in I/B/E/S from January 1995 until June 2006. In event time, day 0 is the day of the announcement. The cumulative abnormal return for each stock is the raw buy-and-hold return adjusted using the estimated beta from market model. The earnings surprise for an announcement is the difference between actual earnings for the quarter recorded by I/B/E/S and the median analyst forecast included in the I/B/E/S detail file during the 30 days before the quarterly earnings announcement scaled by the stock price 5 trading days before the announcement. Quantiles 1 to 5 contain earnings announcements for five quintiles of negative earnings surprises and quantiles 7 to 11 contain earnings surprises for five quintiles of positive earnings surprises. Quantile 6 contains all announcements with an earnings surprise equal to zero. Since the number of negative earnings surprises, positive earnings surprises, and surprises equal to zero are unequal, the number of observations in each quantile differ. In some specifications we follow a similar procedure but sort announcements into surprise deciles instead of quantiles. The breakpoints for the quantiles (deciles) are determined separately for each year. The coefficients in the table are evaluated at the average value of the controls. Columns 1, 2, and 3 analyze observations from the top quantile (11) and the bottom quantile (1). Column 4 analyzes observations from the top decile (10) and bottom decile (1). Columns 5 and 6 analyze observations from the top two quantiles (10 and 11) or the bottom two quantiles (1 and 2). Column 7 analyzes observations from the top two deciles (9 and 10) and bottom two deciles (1 and 2). The standard set of controls includes indicators for the year of announcement, the month of announcement, and the decile of a firm's market capitalization. The earnings surprise volatility controls in columns 3, 4, 6, and 7 are indicators for the decile of the company's earnings surprise standard deviation during the previous 4 years (requiring at least four observations). If a control is included in a regression, that control is also interacted with the indicator variable for the top quantile (or top two quantiles). Announcements made on Saturday or Sunday are excluded from the sample. Standard errors adjusted for heteroskedasticity and clustered by day of announcement are in parentheses (* significant at 10%," significant at 5%, *** significant at 1%). Panel A: The Dependent Variable Is the Cumulative Abnormal Return in Event Time from 0 to 1 Constant Friday Top Group (Top Group) * Friday Top Two Groups (Top Two Groups) + Friday (1) -0.0369 (2) -0.0370 (3) -0.0340 (4) -0.0330 (5) -0.0334 (6) (7) -0.0322 (0.0013)*** (0.0024)*** (0.0019)*** (0.0011)*** 0.0021 0.0003 0.0015 (0.0040) (0.0040) (0.0043) 0.0010 (0.0031) (0.0009)*** 0.0024 (0.0027) (0.0012)*** 0.0012 (0.0029) -0.0254 (0.0007)*** 0.0008 (0.0022) 0.0659 0.0634 0.0640 (0.0014)*** (0.0015)*** (0.0021)*** 0.0635 (0.0014)*** -0.0081 -0.0059 -0.0090 -0.0071 (0.0047)* (0.0048) (0.0050)* (0.0044) Standard Controls (Interacted) Surprise Volatility Controls (Interacted) X X Sorting Procedure 11 Quantiles 11 Quantiles 11 Quantiles R N 0.0915 N = 24,874 0.0990 0.1049 N = 24,874 N = 19,888 X X 10 Deciles 0.1148 N = 23,458 0.0582 (0.0010)*** -0.0092 (0.0031)*** 11 Quantiles 0.0864 N = 49,538 0.0571 (0.0013)*** -0.0086 (0.0034)*** X 11 Quantiles 0.0985 N = 40,331 0.0518 (0.0009)*** -0.0095 (0.0028)*** X X 10 Deciles 0.0986 N = 47,879 (continued)