About kelSAS

kelSAS · ‎03-03-2020

I apologize if this looks repetitive, but I was told to post it here. Two equations Y = X + V1 + V2 + industry + time X = Y + V1 + V2 + industry + time I have an instrument, Z, that can be used to predict X. I also have about 100 of industry and time dummies. (So I can't use 'model'). How do I go about producing the estimated coefficients for the above two equations? No STATA, nor SAS seem to have the procedure for simultaneous equations.

kelSAS · ‎03-02-2020

Two equations Y = X + V1 + V2 + industry + time X = Y + V1 + V2 + industry + time I have an instrument, Z, that can be used to predict X. I also have about 100 of industry and time dummies. (So I can't use 'model'). How do I go about producing the estimated coefficients for the above two equations? No STATA, nor SAS seem to have the procedure for simultaneous equations.

kelSAS · ‎03-02-2020

Thanks, but this only handles one group. I have about 300 groups.

kelSAS · ‎02-22-2020

I want to compare two coefficients of the same model with two different sample. To do so, I need to interact all variables in the model with the dummy variable that divides total observations into the two different samples. I want to display tables that have the coefficients on all variables with one sample and the same with the other sample, and the third column would show the difference between the two samples. Because the fixed effects are going to change the coefficients if I run the two samples separately, I need to have all variables interacted with the indicator variable and include in the model. Can you tell me how to create the loop, please?

kelSAS · ‎02-22-2020

I have generated dummy variables for every year as below Data work; set work; array yrdummy(*) d_&startyr - d_&endyr; do i=1 to dim(yrdummy); if fyear = i+&startyr then yrdummy{i} = 1; else yrdummy{i} = 0; end; drop i; run; Now I want to interact every d_'year' with a variable. For example, I have a dummy called 'd_group'. I want to interact d_group with d_1988 through d_2018 so that I am left with variables called d_group*d_1988 through d_group*d_2018. Is there a loop to make this happen?

kelSAS · ‎02-12-2020

I have firm-year observations with industry data. Using this data, I have created industry*year interaction called 'group1'. I would like to 1) cluster standard error by firm-year 2) include firm and group1 fixed effect What I could have used (which worked if the fixed effects were only industry and year) is as below: proc surveyreg data=work; cluster fyear gvkey; class gvkey group1; model dv=iv1 iv2 iv3 gvkey group1 / adjrsq solution; quit; This no longer works because I have too many variables for fixed effects (both firm and group1). I also cannot do 2-way clustering (fyear and gvkey) in this procedure. I have been downloading the dataset and run the tests with STATA, but I believe there must be a way to yield the same results using SAS. Would anyone be able to help me with any suggestions? Thank you in advance.

kelSAS · ‎02-12-2020

Thank you so much. You saved lots of space in the disk and time for me!

kelSAS · ‎02-12-2020

I am using SAS on my computer to log on a remote server and run my codes in batches. I had always included a statement looking like below. %include "c:\Users\Me\macro1.sas"; Then, whenever I batch submit, SAS would run without telling me that it successfully loaded all the macros. All of sudden, I am getting all of my macro code written on my log file with comments like below NOTE: %INCLUDE (level 1) ending. 2589 + 2590 +%let dropvar = ; 2591 +%if &byvar = none %then %do; 2592 + 2593 +data xtemp; 2594 +set xtemp; I have about 20 macros to load everytime I batch submit, but about 5 of them are giving me the log file. Would anyone possibly know the cause and the solution not to have these anymore? For one line of code, I am having more than 6,000 lines of log. Thank you for your help in advance.

kelSAS · ‎06-26-2018

I apologize that my explanation was inadequate. FE is for fixed effect. So original model looks like below. y = α1 x1 + α2 D + α3 x1*D + α4 d_1990 + α5 d_1991 + ... + d_2017 + d_sic1 + d_sic2 + ... + d_sic48 + error where d_1990 to d_2017 are year dummies and d_sic1 to d_sic48 are industry dummies. What I want is to look at the difference in α3 for the sample with Z=0 and the sample with Z=1. But because of FE (fixed effects) that are laid out as d_1990 to d_2017 and d_sic1 to d_sic48, if I run the regression separately for Z=0 group and Z=1 group, the the difference in α3 from the two regressions is not the same as the coefficient on 3-way interaction (x1*D*Z) from 1 multiple regression with bunch of interactions. I know the intuition and interpretation won't be different, but I need the numbers (the difference and the 3-way interaction) to be the same to put in my result table.

kelSAS · ‎06-26-2018

I have a multiple regression equation with year and industry fixed effect as below. y = x1 + D + x1*D + FE + error (x1 is continuous and D is binary) and I have another dummy variable (let's say, Z = 0 or 1). What I want to see is if the coefficients on x1*D are statistically significantly different for the subsample Z =1 and Z=0. What I have tried is 3-way interaction (x1*D*Z). However, because of fixed effects, the difference in the coefficient on x1*D from separate regression for Z=0 and Z=1 is not the same as the coefficient on 3-way interaction. (Separate regression allows the error term to vary separately whereas 3-way interaction multiple regression does not) In the end, what I want is 3 columns with the coefficients for Z=0, Z=1, and the difference. Is there some type of test in SAS where I can achieve this?

kelSAS · ‎06-14-2018

Thank you so much for your active input. I wish there was one step, but your two-step works perfectly.

kelSAS · ‎06-14-2018

Thank you for asking that question. I need to handle outliers properly and when I run 'proc univariate', the output would show the maximum and minimum with observation number and the value of the variable I ran 'proc univariate' with. I could use certain value to limit data that's proc print'ed, but was wondering if I could print several datalines by observation number in one proc print step.

kelSAS · ‎06-13-2018

Yes! I want to see only those observations printed.

kelSAS · ‎06-13-2018

I understand using firstobs=x, and obs=x, to print out all variables for that observations. I am trying to use proc print for several observations that are far from each other in observation numbers. So, for example, if I want observations number 38701, I would do proc print data=dataset (firstobs=38701, obs=38701). Would anyone know how to print two or three observations that have observations numbers like 38701, 14080, and/or 9631. Thank you for your time in advance.

kelSAS · ‎05-06-2018

I am trying to run regression of GDP on Investment by year, and want to t-test the coefficient on investment by dividing the sample into two groups in the middle year of all years. For this, I had the following. proc reg data=work noprint outest=investdata; by year; model gdp=investm1 invest investp1; run; investm1 and investp1 are lag and lead values of invest because I needed to control them in the regression. Then for t-test, I split the years into g=0 or 1 at the mid-year of sample years using data step on dataset 'investdata'. Then I typed in proc ttest data=investdata; class g; var invest; run; I get error message saying I don't have n,std,mean. Desired output would be whether coefficient on invest is significantly different from before and after at the five percent significance level. Where did I do wrong? Is there other ways I could possibly get the output that I want? Thank you for your help in advance.

Online Status	Offline
Date Last Visited	‎01-27-2022 02:44 PM

Simultaneous equations

Simultaneous equations

Re: multiple Interaction

Re: multiple Interaction

multiple Interaction

The ultimate fixed effect with clustering: proc surveyreg

Re: how to exclude %include comment

how to exclude %include comment

Re: Statistical test for the difference between coefficients in multip...

Statistical test for the difference between coefficients in multiple r...

proc print a few observations

Simultaneous equations

Simultaneous equations

Re: multiple Interaction

Re: multiple Interaction

multiple Interaction

The ultimate fixed effect with clustering: proc surveyreg

Re: how to exclude %include comment

how to exclude %include comment

Re: Statistical test for the difference between coefficients in multip...

Statistical test for the difference between coefficients in multiple r...

Re: proc print a few observations

Re: proc print a few observations

Re: proc print a few observations

proc print a few observations

t-test after regression by year