1. (StataNow™) DO-FILE EDITOR TEMPLATES. By popular request, you can now save time and ensure consistency when you can create new documents in the Do-file Editor by using Stata templates and user-defined templates. You can access templates by using the New > Document from template menu. Selecting an item from the templates menu will open an editor and set the contents of the editor to the contents of the template. You can define your own templates by creating the document_templates directory in your PERSONAL directory and then saving template files into that directory. See [GSW] 13 Using the Do-file Editor—automating Stata (or [GSM] 13 Using the Do-file Editor—automating Stata or [GSU] 13 Using the Do-file Editor—automating Stata) to learn more.
2. (StataNow™) ENHANCEMENTS FOR LISTING DATA. Two new options are added to list:
   1. Option sepbyexp(exp) draws a separator line whenever the value of expression exp changes. exp does not necessarily have to refer to variables in the dataset.
   2. Option footer displays variable names as a footer.
3. xthdidregress and hdidregress have new option basetime(), which determines the base time to be used for the pretreatment periods when computing ATETs. Option basetime(adaptive), the default, specifies that the base time for pretreatment periods is adaptive. That is, for each cohort g at time t, the base time is t – 1 in the pretreatment periods and g – 1 in the posttreatment periods. Option basetime(common) specifies that all periods, both pretreatment and posttreatment, use the common base time g – 1 for each cohort g. The common base time is particularly useful when evaluating whether the assumption of parallel trends is violated in event studies.
4. regress has new stored result e(sum_w), containing the sum of the weights.
5. The Java Development Kit (OpenJDK) that is redistributed with Stata is now updated to version 17.0.13+11-LTS acquired from Azul Systems.

© Copyright 1996–2026 StataCorp LLC. All rights reserved.

6. exlogistic and expoisson have the following fixes:
   1. exlogistic and expoisson produced a coefficient table that did not use the display formats set by set cformat, set pformat, or set sformat. This has been fixed.
   2. exlogistic, test(score) and expoisson, irr correctly produced results with the exponentiated CI upper bound, but stored result r(table), containing the coefficient-table matrix, did not exponentiate the CI upper bound. This has been fixed.
7. lpirf, when used with panel data, ignored the panel structure, treating it as one long time series, and presented incorrect results. This has been fixed, and lpirf now produces an error message when used with panel data.
8. margins has the following fixes:
   1. margins, when specified with option asbalanced or atmeans and a time-series-operated factor-variable (tsfvar) in the marginlist, exited with error message “factor tsfvar not found in list of covariates”, even when tsfvar was present in the list of covariates. This has been fixed.
   2. margins with one variable specified in marginslist and option atmeans or asbalanced displayed a header erroneously describing the variable in marginslist as being set to its mean or treated as balanced. This has been fixed.
9. sem with option method(mlmv), when encountering difficulty computing the log likelihood for observations within a missing value pattern of the observed variables, sometimes exited with an unhelpful error message. In such cases, the iteration log typically contains “(backed up)” or “(not concave)” messages. The unhelpful error message is no longer presented, and iterations continue until the model either converges or reaches the maximum number of iterations.
10. In rare cases, Stata could crash when executing programs or do-files if the Data Editor was open. This has been fixed.