Regression outputs

[Parvesh Seeballack]

Hi Stata Users

I have been working on a event study which involved the following two phases:

1) Calculating the AR and CAR for a number of event windows
2) Regressing my CAR on a number of variables

I have appended the CAR to a file using the following:

keep in 1/1
append using 6day_car
save 6day_car, replace
restore

My question is how to I export the results of my regressions from (1) to an excel / word file where I am left with only the AR / CAR, t-statistics and the significance levels.

I would like to have the same for part (2) where I am left with the beta, t-statistics and the significance levels. I have the used the following code to get my results to excel:

forvalues i = 1/1324 {
preserve
keep if obs == `i'
eststo: quietly regress car `var'
esttab using regresults.csv, replace
restore
}

My problem is that it gives me the following error:

matsize too small
You have attempted to create a matrix with too
many rows or columns or attempted to fit a model
with too many variables. You need to increase
matsize; it is currently 400. Use set matsize;
see help matsize.

If you are using factor variables and included an
interaction that has lots of missing cells, either
increase matsize or set emptycells drop to reduce
the required matrix size; see help set emptycells.

If you are using factor variables, you might have
accidentally treated a continuous variable as a
categorical, resulting in lots of categories. Use
the c. operator on such variables.

I have not been able to find a way to get all results on excel since.

Would really appreciate if someone could help me. Its the final stage of my work...exporting the file, building my tables and explain the results.

Thank you guys in advance for your help!

[Phil Bromiley]

You didn't get a quick answer. You have a difficult question to follow - following the FAQ on asking question might help.

First, you seem to want to keep some material from estimating the CARs. That's fine, but we'd need to see the code that did the CAR estimates to suggest how you'd keep what you want to keep. You can generally export results using estout or outreg2 - they have many options.

You need to identify where the problem occurs. Entering set trace on before the loop will help you understand where exactly your "matsize too small" error is appearing. Normally, this occurs when you have piles of variables. The simplest thing would be to increase the matsize. Did you try this?

I also wonder about
esttab using regresults.csv, replace Why wouldn't this just replace the file every time it runs? Don't you want to append? Also, it is seldom good practice to run a pile of regressions with one dv and differing iv's.

[Parvesh Seeballack]

Hi Phil

Thank you for replying.

Well here is my code for calculating the CAR:

use eventsdates, clear
drop date_id
forvalues i = 1/1324 {
preserve
keep if obs == `i'
joinby nnn using marketdata
sort date
drop date_id
g date_id = _n
gen day_cnt = date_id
gen target_day = day_cnt if date==event_date
egen max_target_day = max(target_day)
gen evday = day_cnt-max_target_day
drop day_cnt target_day max_target_day
sort evday
gen evt_window=1 if evday>=0 & evday<=6
gen est_window=1 if evday<=-11 & evday>=-30
drop if evt_window==. & est_window==.
foreach var of local vars {
reg return_`var' returnmarket if est_window==1
gen rmse_`var' = e(rmse)
predict phat_`var'
gen ar_`var' = return_`var' - phat_`var' if evt_window==1
drop phat_`var'
}
drop if evt_window==.
drop est_window nnn
***************************************************
*Display the CAR and its Test Statistic
foreach var of local vars {
egen car_`var' = sum(ar_`var')
gen tstat_`var' = car_`var'/(rmse_`var'*sqrt(_N))
drop return_`var' rmse_`var' ar_`var'
}
drop returnmarket date_id evday evt_window date
* DO EVENT analysis, generate CAR in 1/1
keep in 1/1
append using 6day_car
save 6day_car, replace
restore
}
use 6day_car, clear

Actually I'd like to be able to send the CAR, tstat and confidence interval, i.e. the stars, on one excel sheet.

[Phil Bromiley]

Some folks like appending. I tend to not do this - a matter of taste.

One way to solve your problem to create four variables with missing values before the - one string and the other three numerical and a counter. Then in the loop, write the values into the appropriate variables in a given observation.

Something like:
g car=.
g tstat=.
g var=" "
local i=1
foreach var of local vars {
su ar_`var'
replace car=r(mean) in `i'/`i'
local i=`i'+1
}
At the end, you can export these variables.

[Parvesh Seeballack]

Hi everyone

I have been able to get my tables in excel with your help!

Thanks so much for the advice guys.

Regards
Parvesh