Lincom - expression too long

[Barbara Tuennes]

Dear Statalist users,

I'm using the lincom command, but apparently including too many coefficients, as I get the error message "expression too long".
I read that with the nlcom command, it's possible to shorten the expression.
However, this does not wok with lincom.

Any help is greatly appreciated.

Best,

Barbara

[Svend Juul]

Please, tell exactly what commands you gave and what Stata responded. "Does not work" is not enough to elicit an informed feedback.

[Barbara Tuennes]

Please excuse my imprecise information. I used the following commands:

/* Regression */
reg retired int193500* int193525* int193550* int193575* ///
int193600* int193625* int193650* ///
int193700* int193725* int193750* int193775* ///
int193800* int193825* int193850* int193875* ///
int193900* int193925* int193950* int193975* ///
int194000* int194025* int194050* int194075* ///
int194100* int194125* int194150* int194175* ///
int194200* int194225* int194250* int194275* ///
int194300* int194325* int194350* int194375* ///
int194400* int194425* int194450* int194475* ///
int194500* int194525* int194550* int194575* ///
int194600* int194625* int194650* int194675* ///
int194700*, noconstant cluster(id)


/* Post-Estimation - very long command with lots of variables, all dummy-variables for birth quarter and "age quarter" */
lincom 1/41 * ///
(- (int1937006000+ int1937006025 + int1937006050 + int1937006075 + ///
int1937006100 + int1937006125 + int1937006150 + int1937006175 + ///
int1937006200 + int1937006225 + int1937006250 + int1937006275 + ///
int1937006300 + int1937006325 + int1937006350 + int1937006375 + ///
int1937006400 + int1937006425 + int1937006450 + int1937006475 + ///
int1937006500 + int1937006525 + int1937006550 + int1937006575) ///
- 1/2* (int1937256000+ int1937256025 + int1937256050 + int1937256075 + ///
int1937256100 + int1937256125 + int1937256150 + int1937256175 + ///
int1937256200 + int1937256225 + int1937256250 + int1937256275 + ///
int1937256300 + int1937256325 + int1937256350 + int1937256375 + ///
int1937256400 + int1937256425 + int1937256450 + int1937256475 + ///
int1937256500 + int1937256525 + int1937256550 + int1937256575) ///
-1/3 * (int1937506000+ int1937506025 + int1937506050 + int1937506075 + ///
int1937506100 + int1937506125 + int1937506150 + int1937506175 + ///
int1937506200 + int1937506225 + int1937506250 + int1937506275 + ///
int1937506300 + int1937506325 + int1937506350 + int1937506375 + ///
int1937506400 + int1937506425 + int1937506450 + int1937506475 + ///
int1937506500 + int1937506525 + int1937506550 + int1937506575) ///
-1/4 * (int1937756000+ int1937756025 + int1937756050 + int1937756075 + ///
int1937756100 + int1937756125 + int1937756150 + int1937756175 + ///
int1937756200 + int1937756225 + int1937756250 + int1937756275 + ///
int1937756300 + int1937756325 + int1937756350 + int1937756375 + ///
int1937756400 + int1937756425 + int1937756450 + int1937756475 + ///
int1937756500 + int1937756525 + int1937756550 + int1937756575) ///
-1/5* (int1938006000+ int1938006025 + int1938006050 + int1938006075 + ///
int1938006100 + int1938006125 + int1938006150 + int1938006175 + ///
int1938006200 + int1938006225 + int1938006250 + int1938006275 + ///
int1938006300 + int1938006325 + int1938006350 + int1938006375 + ///
int1938006400 + int1938006425 + int1938006450 + int1938006475 + ///
int1938006500 + int1938006525 + int1938006550 + int1938006575) ///
- 1/6* (int1938256000+ int1938256025 + int1938256050 + int1938256075 + ///
int1938256100 + int1938256125 + int1938256150 + int1938256175 + ///
int1938256200 + int1938256225 + int1938256250 + int1938256275 + ///
int1938256300 + int1938256325 + int1938256350 + int1938256375 + ///
int1938256400 + int1938256425 + int1938256450 + int1938256475 + ///
int1938256500 + int1938256525 + int1938256550 + int1938256575) ///
-1/7 * (int1938506000+ int1938506025 + int1938506050 + int1938506075 + ///
int1938506100 + int1938506125 + int1938506150 + int1938506175 + ///
int1938506200 + int1938506225 + int1938506250 + int1938506275 + ///
int1938506300 + int1938506325 + int1938506350 + int1938506375 + ///
int1938506400 + int1938506425 + int1938506450 + int1938506475 + ///
int1938506500 + int1938506525 + int1938506550 + int1938506575) ///
-1/8 * (int1938756000+ int1938756025 + int1938756050 + int1938756075 + ///
int1938756100 + int1938756125 + int1938756150 + int1938756175 + ///
int1938756200 + int1938756225 + int1938756250 + int1938756275 + ///
int1938756300 + int1938756325 + int1938756350 + int1938756375 + ///
int1938756400 + int1938756425 + int1938756450 + int1938756475 + ///
int1938756500 + int1938756525 + int1938756550 + int1938756575) ///
-1/9* (int1939006000+ int1939006025 + int1939006050 + int1939006075 + ///
int1939006100 + int1939006125 + int1939006150 + int1939006175 + ///
int1939006200 + int1939006225 + int1939006250 + int1939006275 + ///
int1939006300 + int1939006325 + int1939006350 + int1939006375 + ///
int1939006400 + int1939006425 + int1939006450 + int1939006475 + ///
int1939006500 + int1939006525 + int1939006550 + int1939006575) ///
- 1/10* (int1939256000+ int1939256025 + int1939256050 + int1939256075 + ///
int1939256100 + int1939256125 + int1939256150 + int1939256175 + ///
int1939256200 + int1939256225 + int1939256250 + int1939256275 + ///
int1939256300 + int1939256325 + int1939256350 + int1939256375 + ///
int1939256400 + int1939256425 + int1939256450 + int1939256475 + ///
int1939256500 + int1939256525 + int1939256550 + int1939256575) ///
-1/11 * (int1939506000+ int1939506025 + int1939506050 + int1939506075 + ///
int1939506100 + int1939506125 + int1939506150 + int1939506175 + ///
int1939506200 + int1939506225 + int1939506250 + int1939506275 + ///
int1939506300 + int1939506325 + int1939506350 + int1939506375 + ///
int1939506400 + int1939506425 + int1939506450 + int1939506475 + ///
int1939506500 + int1939506525 + int1939506550 + int1939506575) ///
-1/12 * (int1939756000+ int1939756025 + int1939756050 + int1939756075 + ///
int1939756100 + int1939756125 + int1939756150 + int1939756175 + ///
int1939756200 + int1939756225 + int1939756250 + int1939756275 + ///
int1939756300 + int1939756325 + int1939756350 + int1939756375 + ///
int1939756400 + int1939756425 + int1939756450 + int1939756475 + ///
int1939756500 + int1939756525 + int1939756550 + int1939756575) ///
-1/13 * (int1940006000+ int1940006025 + int1940006050 + int1940006075 + ///
int1940006100 + int1940006125 + int1940006150 + int1940006175 + ///
int1940006200 + int1940006225 + int1940006250 + int1940006275 + ///
int1940006300 + int1940006325 + int1940006350 + int1940006375 + ///
int1940006400 + int1940006425 + int1940006450 + int1940006475 + ///
int1940006500 + int1940006525 + int1940006550 + int1940006575) ///
- 1/14* (int1940256000+ int1940256025 + int1940256050 + int1940256075 + ///
int1940256100 + int1940256125 + int1940256150 + int1940256175 + ///
int1940256200 + int1940256225 + int1940256250 + int1940256275 + ///
int1940256300 + int1940256325 + int1940256350 + int1940256375 + ///
int1940256400 + int1940256425 + int1940256450 + int1940256475 + ///
int1940256500 + int1940256525 + int1940256550 + int1940256575) ///
-1/15 * (int1940506000+ int1940506025 + int1940506050 + int1940506075 + ///
int1940506100 + int1940506125 + int1940506150 + int1940506175 + ///
int1940506200 + int1940506225 + int1940506250 + int1940506275 + ///
int1940506300 + int1940506325 + int1940506350 + int1940506375 + ///
int1940506400 + int1940506425 + int1940506450 + int1940506475 + ///
int1940506500 + int1940506525 + int1940506550 + int1940506575) ///
-1/16 * (int1940756000+ int1940756025 + int1940756050 + int1940756075 + ///
int1940756100 + int1940756125 + int1940756150 + int1940756175 + ///
int1940756200 + int1940756225 + int1940756250 + int1940756275 + ///
int1940756300 + int1940756325 + int1940756350 + int1940756375 + ///
int1940756400 + int1940756425 + int1940756450 + int1940756475 + ///
int1940756500 + int1940756525 + int1940756550 + int1940756575) ///
- 1/17 * (int1941006000+ int1941006025 + int1941006050 + int1941006075 + ///
int1941006100 + int1941006125 + int1941006150 + int1941006175 + ///
int1941006200 + int1941006225 + int1941006250 + int1941006275 + ///
int1941006300 + int1941006325 + int1941006350 + int1941006375 + ///
int1941006400 + int1941006425 + int1941006450 + int1941006475 + ///
int1941006500 + int1941006525 + int1941006550 + int1941006575) ///
- 1/18* (int1941256000+ int1941256025 + int1941256050 + int1941256075 + ///
int1941256100 + int1941256125 + int1941256150 + int1941256175 + ///
int1941256200 + int1941256225 + int1941256250 + int1941256275 + ///
int1941256300 + int1941256325 + int1941256350 + int1941256375 + ///
int1941256400 + int1941256425 + int1941256450 + int1941256475 + ///
int1941256500 + int1941256525 + int1941256550 + int1941256575) ///
-1/19 * (int1941506000+ int1941506025 + int1941506050 + int1941506075 + ///
int1941506100 + int1941506125 + int1941506150 + int1941506175 + ///
int1941506200 + int1941506225 + int1941506250 + int1941506275 + ///
int1941506300 + int1941506325 + int1941506350 + int1941506375 + ///
int1941506400 + int1941506425 + int1941506450 + int1941506475 + ///
int1941506500 + int1941506525 + int1941506550 + int1941506575) ///
-1/20 * (int1941756000+ int1941756025 + int1941756050 + int1941756075 + ///
int1941756100 + int1941756125 + int1941756150 + int1941756175 + ///
int1941756200 + int1941756225 + int1941756250 + int1941756275 + ///
int1941756300 + int1941756325 + int1941756350 + int1941756375 + ///
int1941756400 + int1941756425 + int1941756450 + int1941756475 + ///
int1941756500 + int1941756525 + int1941756550 + int1941756575) ///
-1/21 * (int1942006000+ int1942006025 + int1942006050 + int1942006075 + ///
int1942006100 + int1942006125 + int1942006150 + int1942006175 + ///
int1942006200 + int1942006225 + int1942006250 + int1942006275 + ///
int1942006300 + int1942006325 + int1942006350 + int1942006375 + ///
int1942006400 + int1942006425 + int1942006450 + int1942006475 + ///
int1942006500 + int1942006525 + int1942006550 + int1942006575) ///
- 1/22* (int1942256000+ int1942256025 + int1942256050 + int1942256075 + ///
int1942256100 + int1942256125 + int1942256150 + int1942256175 + ///
int1942256200 + int1942256225 + int1942256250 + int1942256275 + ///
int1942256300 + int1942256325 + int1942256350 + int1942256375 + ///
int1942256400 + int1942256425 + int1942256450 + int1942256475 + ///
int1942256500 + int1942256525 + int1942256550 + int1942256575) ///
-1/23 * (int1942506000+ int1942506025 + int1942506050 + int1942506075 + ///
int1942506100 + int1942506125 + int1942506150 + int1942506175 + ///
int1942506200 + int1942506225 + int1942506250 + int1942506275 + ///
int1942506300 + int1942506325 + int1942506350 + int1942506375 + ///
int1942506400 + int1942506425 + int1942506450 + int1942506475 + ///
int1942506500 + int1942506525 + int1942506550 + int1942506575) ///
-1/24 * (int1942756000+ int1942756025 + int1942756050 + int1942756075 + ///
int1942756100 + int1942756125 + int1942756150 + int1942756175 + ///
int1942756200 + int1942756225 + int1942756250 + int1942756275 + ///
int1942756300 + int1942756325 + int1942756350 + int1942756375 + ///
int1942756400 + int1942756425 + int1942756450 + int1942756475 + ///
int1942756500 + int1942756525 + int1942756550 + int1942756575) ///
-1/25* (int1943006000+ int1943006025 + int1943006050 + int1943006075 + ///
int1943006100 + int1943006125 + int1943006150 + int1943006175 + ///
int1943006200 + int1943006225 + int1943006250 + int1943006275 + ///
int1943006300 + int1943006325 + int1943006350 + int1943006375 + ///
int1943006400 + int1943006425 + int1943006450 + int1943006475 + ///
int1943006500 + int1943006525 + int1943006550 + int1943006575) ///
- 1/26* (int1943256000+ int1943256025 + int1943256050 + int1943256075 + ///
int1943256100 + int1943256125 + int1943256150 + int1943256175 + ///
int1943256200 + int1943256225 + int1943256250 + int1943256275 + ///
int1943256300 + int1943256325 + int1943256350 + int1943256375 + ///
int1943256400 + int1943256425 + int1943256450 + int1943256475 + ///
int1943256500 + int1943256525 + int1943256550 + int1943256575) ///
-1/27 * (int1943506000+ int1943506025 + int1943506050 + int1943506075 + ///
int1943506100 + int1943506125 + int1943506150 + int1943506175 + ///
int1943506200 + int1943506225 + int1943506250 + int1943506275 + ///
int1943506300 + int1943506325 + int1943506350 + int1943506375 + ///
int1943506400 + int1943506425 + int1943506450 + int1943506475 + ///
int1943506500 + int1943506525 + int1943506550 + int1943506575) ///
-1/28 * (int1943756000+ int1943756025 + int1943756050 + int1943756075 + ///
int1943756100 + int1943756125 + int1943756150 + int1943756175 + ///
int1943756200 + int1943756225 + int1943756250 + int1943756275 + ///
int1943756300 + int1943756325 + int1943756350 + int1943756375 + ///
int1943756400 + int1943756425 + int1943756450 + int1943756475 + ///
int1943756500 + int1943756525 + int1943756550 + int1943756575) ///
-1/29 * (int1944006000+ int1944006025 + int1944006050 + int1944006075 + ///
int1944006100 + int1944006125 + int1944006150 + int1944006175 + ///
int1944006200 + int1944006225 + int1944006250 + int1944006275 + ///
int1944006300 + int1944006325 + int1944006350 + int1944006375 + ///
int1944006400 + int1944006425 + int1944006450 + int1944006475 + ///
int1944006500 + int1944006525 + int1944006550 + int1944006575) ///
- 1/30* (int1944256000+ int1944256025 + int1944256050 + int1944256075 + ///
int1944256100 + int1944256125 + int1944256150 + int1944256175 + ///
int1944256200 + int1944256225 + int1944256250 + int1944256275 + ///
int1944256300 + int1944256325 + int1944256350 + int1944256375 + ///
int1944256400 + int1944256425 + int1944256450 + int1944256475 + ///
int1944256500 + int1944256525 + int1944256550 + int1944256575) ///
-1/31 * (int1944506000+ int1944506025 + int1944506050 + int1944506075 + ///
int1944506100 + int1944506125 + int1944506150 + int1944506175 + ///
int1944506200 + int1944506225 + int1944506250 + int1944506275 + ///
int1944506300 + int1944506325 + int1944506350 + int1944506375 + ///
int1944506400 + int1944506425 + int1944506450 + int1944506475 + ///
int1944506500 + int1944506525 + int1944506550 + int1944506575) ///
-1/32 * (int1944756000+ int1944756025 + int1944756050 + int1944756075 + ///
int1944756100 + int1944756125 + int1944756150 + int1944756175 + ///
int1944756200 + int1944756225 + int1944756250 + int1944756275 + ///
int1944756300 + int1944756325 + int1944756350 + int1944756375 + ///
int1944756400 + int1944756425 + int1944756450 + int1944756475 + ///
int1944756500 + int1944756525 + int1944756550 + int1944756575) ///
-1/33* (int1945006000+ int1945006025 + int1945006050 + int1945006075 + ///
int1945006100 + int1945006125 + int1945006150 + int1945006175 + ///
int1945006200 + int1945006225 + int1945006250 + int1945006275 + ///
int1945006300 + int1945006325 + int1945006350 + int1945006375 + ///
int1945006400 + int1945006425 + int1945006450 + int1945006475 + ///
int1945006500 + int1945006525 + int1945006550 + int1945006575))

/* the error message I get is:*/

expression too long
r(130);

[Carlo Lazzaro]

Barbara:
are you sure that you need 48 predictors in your regression?
Is there no other way to specify the regression model and reduce the number of predictors?

Kind regards,
Carlo

[Barbara Tuennes]

Dear Carlo,
yes, I'm unfortunately very sure that I need this large amount of predictors. I have to estimate this as an alternative to a simpler model with less predictors.
Best,
Barbara

[Carlo Lazzaro]

Barbara:
if your goal, as I understand it, is to compare two linear regression models with the same dependent variable (retired, provided that this is a continuos variable) but different amount of predictors, you can possibly avoid -lincom- and go on like this:

reg retired x1 ...x1000 /// this is the regression model with the highest number of predictors: let's call it full
estimate store full
reg retired x1 ...x100 /// this is the regression model with the lowest number of predictors: let's call it minimal
estimate store minimal
estimate stats full minimal

This code will you give back a table displaying Akaike's information criterion (AIC) and Bayesian information criterion (BIC) (for both AIC and BIC the rule is: the lower, the better).

Kind regards,
Carlo

[Attaullah Shah]

Barbara, if you are interested in comparing coefficients comparison and their marginal effects, you can use z-score. For example, you can run both the regressions with a given variable, and then

z-score = (B1-B2)/sqrt(seB1^2+seB2^)

[Rich Goldstein]

I am not clear why you want to compare the two estimates or what you want to say about this comparison, but you might have a look at -sureg- and -suest-; if, on the other hand, you want to compare predicted values, I would suggest getting the predictions from each model and comparing those (either with a test or graphically)

[Maarten Buis]

z-score = (B1-B2)/sqrt(seB1^2+seB^2)

That formula is incorrect. The problem is that the standard error of the difference between two coefficients should also include the covariance and not just the two variances. The correct formula is:

\[
z = \frac{\beta_1 - \beta_2}{\sqrt{se_{\beta_1}^2 + se_{\beta_2}^2 -2\mathrm{cov}(\beta_1,\beta_2)}}
\]

[Attaullah Shah]

Thanks Maarten Buis for the correciton. Can you refer me to some reading material on this. I had seen this in one of the statalist archive, see

http://www.stata.com/statalist/archive/2012-08/msg00120.html

[Attaullah Shah]

Maarten Buis And please also have a loot at this linkhttp://www.stata.com/statalist/archive/2012-08/msg00121.html

[Barbara Tuennes]

Thanks for all the helpful comments, but again, I haven't explained myself precisely enough. Sorry for that.

I said that I wanted to compare two models to underline that there is no way of reducing the number of predictors. Nevertheless, I need "lincom" to sum up different coefficients (I also need the p-values).

So whether I compare models or not is not really an issue. my actual problem evolves around the fact that I cannot use lincom with the extensive number of (necessary) predictors I got.

[Carlo Lazzaro]

Barbara:
I'm afraid I have run short of advices!
I do hope that someone else on the list can help you out substantively.
KInd regards,
Carlo

[Andrew Musau]

See https://www.statalist.org/forums/for...ssion-too-long on the possibility to combine lincom results.