Hi, all, dear SAS community,    Please bear with me if I returned with too simple a stats question.  I ran two Poisson models, one simply with controls, the other with controls plus variables of interest.  /** Model (1) **/ proc genmod data = dataset; model Y = X1 X2 X3 X4 X5 year1-year9 ind1-ind63 / dist=poisson link=log; run; /** Model (2) **/ proc genmod data = dataset; model Y = X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 year1-year9 ind1-ind63 / dist=poisson link=log; run;  I want to check whether one model is more significant (in terms of fit) than the other. In other words, I want to perform a model test like F-Test when comparing OLS models. According to my internet search, Poisson works differently; they suggest using CONTRAST option in PROC GENMOD but I am not really sure. And the ways they show how to use CONTRAST options look so complicated. Please help! I really appreciate you, SAS community!    Sincerely,  KS -,    ... View more

Hi, greetings, My SAS community!  I have been helped so much by the Community. Even though I find no ways to directly return my thanks, I always cherish my thanks to you.    I have a kind of annual firm financial data (panel) over 1999~2022.    Based on this data set, I want to pick a random year between 1999 and 2022, fifty times. Each time, I want to further random-sample 500 obs. prior to 1 + the year randomly picked. Below are the SAS codes I created:  %macro simul(rep=); %do i=1 %to &rep; DATA Compustat_6; SET Compustat_5; yr=rand('integer',1999,2022); run; DATA Survivors; SET Compustat_6; if fyear<yr+1; RUN; proc surveyselect data=Survivors method=srs rep=1 sampsize = 500 seed = 12345 out=Reg_surv; id _all_; run; proc reg data= Reg_surv outest=result_reg_surv noprint; model Tobinq = roa blev; run; proc append base= final_reg_surv data= result_reg_surv force; %end; %mend; %simul(rep=50); I have three questions about the results:  (1) The dataset  result_reg_surv   shows the intercept, coefficients of roa and blev. But I want to add standard error or t-values for the dependent variables, as well as their confidence intervals from each regression. What options should I enter where?  (2) I want to append the simulated regression results onto the dataset  final_reg_surv  iteratively. But SAS reports an error message that the variables Intercept, roa, and blev are not found in  final_reg_surv  . What's wrong with my coding?  (3) After I executed the codes, I opened the dataset  Reg_surv  to check the data on which the 50th regression was made. As I checked the variable yr, yr comprises various years. What I expected was that yr should be a single year value from the last 50th  yr=rand('integer',1999,2022)  random pick-up.    How should I fix these problems based on the codes I created?    Thanks in advance, happy X-mas!    Sincerely,  KS -                        ... View more

Hi, SAS community!    Urgently I need your help.  I have stock data with firm identifier (=CUSIP), stock date (=Date), and daily returns (=return).  d indicates an event day. data have; input CUSIP $ DATE :mmddyy10. return d; format date mmddyy10.; datalines; 0001241 6/28/1988 0.003 . 0001241 6/29/1988 0 . 0001241 6/30/1988 0 . 0001241 7/1/1988 0.018 . 0001241 7/4/1988 0.007 . 0001241 7/5/1988 0.003 . 0001241 7/6/1988 0.010 . 0001241 7/7/1988 0.031 . 0001241 7/8/1988 0.003 . 0001241 7/11/1988 0 . 0001241 7/12/1988 -0.017 . 0001241 7/13/1988 -0.006 . 0001241 7/14/1988 -0.027 . 0001241 7/15/1988 0.003 . 0001241 7/18/1988 -0.028 . 0001241 7/19/1988 0.011 . 0001241 7/20/1988 0 . 0001241 7/21/1988 0 . 0001241 7/22/1988 -0.010 . 0001241 7/25/1988 0.011 . 0001241 7/26/1988 0.010 . 0001241 7/27/1988 0 . 0001241 7/28/1988 0.010 . 0001241 7/29/1988 0.017 . 0001241 8/1/1988 0 . 0001241 8/2/1988 0 . 0001241 8/3/1988 -0.010 . 0001241 8/4/1988 -0.007 . 0001241 8/5/1988 0 . 0001241 8/8/1988 0 . 0001241 8/9/1988 -0.003 . 0001241 8/10/1988 -0.014 . 0001241 8/11/1988 0 . 0001241 8/12/1988 0 . 0001241 8/15/1988 -0.003 . 0001241 8/16/1988 -0.007 . 0001241 8/17/1988 0 . 0001241 8/18/1988 0 . 0001241 8/19/1988 0 . 0001241 8/22/1988 0 . 0001241 8/23/1988 -0.003 . 0001241 8/24/1988 -0.003 . 0001241 8/25/1988 -0.011 . 0001241 8/26/1988 -0.026 . 0001241 8/29/1988 0 . 0001241 8/30/1988 -0.015 . 0001241 8/31/1988 0.015 . 0001241 9/1/1988 -0.003 . 0001241 9/2/1988 0.007 . 0001241 9/5/1988 0.003 . 0001241 9/6/1988 0.003 . 0001241 9/7/1988 0 . 0001241 9/8/1988 -0.015 . 0001241 9/9/1988 -0.007 . 0001241 9/12/1988 0.003 . 0001241 9/13/1988 0.003 . 0001241 9/14/1988 0 . 0001241 9/15/1988 0.023 . 0001241 9/16/1988 -0.018 . 0001241 9/19/1988 -0.015 . 0001241 9/20/1988 -0.011 . 0001241 9/21/1988 0.003 . 0001241 9/22/1988 -0.007 . 0001241 9/23/1988 0.003 . 0001241 9/26/1988 -0.003 . 0001241 9/27/1988 0 . 0001241 9/28/1988 0 . 0001241 9/29/1988 0.003 . 0001241 9/30/1988 0.003 . 0001241 10/3/1988 -0.003 . 0001241 10/4/1988 -0.007 . 0001241 10/5/1988 0.003 . 0001241 10/6/1988 0 . 0001241 10/7/1988 0.003 . 0001241 10/10/1988 0 . 0001241 10/11/1988 0 . 0001241 10/12/1988 -0.003 . 0001241 10/13/1988 0 . 0001241 10/14/1988 0 . 0001241 10/17/1988 0 . 0001241 10/18/1988 0 . 0001241 10/19/1988 0.015 . 0001241 10/20/1988 -0.003 . 0001241 10/21/1988 0.003 . 0001241 10/24/1988 0.011 . 0001241 10/25/1988 0.015 . 0001241 10/26/1988 0.019 . 0001241 10/27/1988 -0.013 . 0001241 10/28/1988 0.005 . 0001241 10/31/1988 0.011 . 0001241 11/1/1988 0.011 . 0001241 11/2/1988 -0.007 . 0001241 11/3/1988 -0.003 . 0001241 11/4/1988 -0.003 . 0001241 11/7/1988 0 . 0001241 11/8/1988 0.003 . 0001241 11/9/1988 -0.003 . 0001241 11/10/1988 0 . 0001241 11/11/1988 0 . 0001241 11/14/1988 0.003 . 0001241 11/15/1988 0.022 . 0001241 11/16/1988 0.043 . 0001241 11/17/1988 0.010 . 0001241 11/18/1988 -0.013 . 0001241 11/21/1988 0 . 0001241 11/22/1988 0 . 0001241 11/23/1988 0 . 0001241 11/24/1988 -0.006 . 0001241 11/25/1988 -0.014 . 0001241 11/28/1988 0 . 0001241 11/29/1988 0 . 0001241 11/30/1988 0.017 . 0001241 12/1/1988 -0.010 . 0001241 12/2/1988 -0.003 . 0001241 12/5/1988 0 . 0001241 12/6/1988 -0.007 . 0001241 12/7/1988 -0.021 . 0001241 12/8/1988 -0.029 . 0001241 12/9/1988 -0.011 . 0001241 12/12/1988 -0.003 . 0001241 12/13/1988 0.003 . 0001241 12/14/1988 0.003 . 0001241 12/15/1988 0 . 0001241 12/16/1988 0 . 0001241 12/19/1988 0 . 0001241 12/20/1988 0.019 . 0001241 12/21/1988 0 . 0001241 12/22/1988 0.003 . 0001241 12/23/1988 0 . 0001241 12/26/1988 0 . 0001241 12/27/1988 0 . 0001241 12/28/1988 0.003 . 0001241 12/29/1988 0.018 . 0001241 12/30/1988 0 . 0001241 1/2/1989 0 . 0001241 1/3/1989 0.029 1 0001241 1/4/1989 0 . 0001241 1/5/1989 0 . 0001241 1/6/1989 0 . 0001241 1/9/1989 0.007 . 0001241 1/10/1989 0.014 . 0001241 1/11/1989 0 . 0001241 1/12/1989 0.006 . 0001241 1/13/1989 0.010 . 0001241 1/16/1989 -0.02 . 0001241 1/17/1989 0 . 0001241 1/18/1989 0.013 . 0001241 1/19/1989 0.013 . 0001241 1/20/1989 0.013 . 0001241 1/23/1989 0.013 . 0001241 1/24/1989 0.016 . 0001241 1/25/1989 0.009 . 0001241 1/26/1989 0.003 . 0001241 1/27/1989 0.025 . 0001241 1/30/1989 0.006 1 0001241 1/31/1989 0 . 0001241 2/1/1989 0 . ;  What I want to achieve is, to extract observations from -120th day up until -21st day from the date designated by d=1.  (Then, I intend calculate means and standard deviations of returns BY CUSIP and 100 days over each [-120,-21]).  There is one condition for what I want to achieve:  In the exemplary dataset above, we have two event days (d=1) on  1/3/1989  and  1/30/1989 .  These two dates are close. Thus, in a dataset of extractions of (-120,-21), many duplicate dates can exist because (-120,-21) from 1/3/1989 and (-120,-21) from 1/30/1989 can share many overlapping days.    For this, I used my SAS codes that I obtained from a kind member of SAS community.  It runs like this:  data want_120_21; start=stop+1; retain stop 0; do stop=stop+1 by 1 until(last.CUSIP); set have end=eof nobs=nobs; by CUSIP; end; put (_n_ nobs sedol start stop ) (=); tradeno=0; do obs=start to stop; set have; if d=1 then do; tradeno+1; tradedate=date; format tradedate date9.; do p=max(obs-120,start) to min(obs-21,stop); offset=p-obs; set have point=p; output; end; end; end; run; I used to use this code to create the windows like (+2,+17), (+2,+126), etc.   But when I apply this code to create the windows (-120,-21), it doesn't work. Keeps producing "Disk is full."    I really appreciate your help,  thanks in advance!!    Sincerely,    KS -  ... View more

Hi, SAS community !    I always appreciate your help-out. I am knocking on your doors to get another little help from you.  Thank you in advance, after all.  The dataset I have looks like this:  data have; input cusip $ year month analysts; datalines; 000255 2004 1 . 000255 2004 2 . 000255 2004 3 1 000255 2004 4 . 000255 2004 5 . 000255 2004 6 . 000255 2004 7 . 000255 2004 8 . 000255 2004 9 . 000255 2004 10 . 000255 2004 11 . 000255 2004 12 . 000255 2005 1 . 000255 2005 2 . 000255 2005 3 . 000255 2005 4 2 000255 2005 5 2 000255 2005 6 . 000255 2005 7 1 000255 2005 8 . 000255 2005 9 . 000255 2005 10 . 000255 2005 11 . 000255 2005 12 . 000307 2015 1 . 000307 2015 2 1 000307 2015 3 1 000307 2015 4 2 000307 2015 5 3 000307 2015 6 . 000307 2015 7 3 000307 2015 8 3 000307 2015 9 . 000307 2015 10 5 000307 2015 11 . 000307 2015 12 . ; run; This data contains the info of how many analysts cover a firm (cusip) in a certain year-month.  One sure thing about this data is that each year has twelve months. I designed it so.    What I want to have is, missing obs. are filled up by the average value of previous and next obs.  Here, firm-year is a chunk. Thus, a firm-year that starts or ends missing should be filled by the first or last non-missing values.    Therefore, I expect to achieve the following:  data have; input cusip $ year month analysts; datalines; 000255 2004 1 1 000255 2004 2 1 000255 2004 3 1 000255 2004 4 1 000255 2004 5 1 000255 2004 6 1 000255 2004 7 1 000255 2004 8 1 000255 2004 9 1 000255 2004 10 1 000255 2004 11 1 000255 2004 12 1 000255 2005 1 2 000255 2005 2 2 000255 2005 3 2 000255 2005 4 2 000255 2005 5 2 000255 2005 6 1.5 000255 2005 7 1 000255 2005 8 1 000255 2005 9 1 000255 2005 10 1 000255 2005 11 1 000255 2005 12 1 000307 2015 1 1 000307 2015 2 1 000307 2015 3 1 000307 2015 4 2 000307 2015 5 3 000307 2015 6 3 000307 2015 7 3 000307 2015 8 3 000307 2015 9 4 000307 2015 10 5 000307 2015 11 5 000307 2015 12 5 ; run; I hope my question were not so stupid.  Thank you all!    Sincerely,    KS -,    ... View more