[Leonardo Guizzetti]

A coalesce() function that operates as follows given a list of variables as arguments, returns the first non missing value in the list, or else missing if all values are missing.

[Adrien Bouguen]

When conducting a search in the do file editor, the items found are highlighted in a light, barely visible, transparent blue which is close to the blue used to display the code. I often found myself spending time looking for the found item. A color that clashes with the code (red? orange?) would be a better idea in my opinion.

[FitzGerald Blindman]

An easy way to export tables after the summarize command (like ourreg2 but not for a regression)

[Scott Adams]

For the -collect- family to allow wildcards to refer to multiple dimensions or levels.

[Lydia Palumbo]

Official command to graph survival curves from Cox w/ time-varying variables

[Erik Ruzek]

Non-linear parameter constraints in sem. See this thread for an example, including the clunky and difficult for non-programmers workaround.

[Niels Henrik Bruun]

A nice feature to add to all regressions is the option adjustments, the argument being a list of variables, factor variables, or interactions. Adjustments should be in the regression model, but their estimates should not be in the regression output except for a listing below the tables.

[Niels Henrik Bruun]

Regarding #127 above. Further, it would be nice, if the adjustments were marked separately in e(b) and e(V).

[Fernando Furquim]

I am sure there are several workarounds, but it would be very convenient to be able to format variables as percentages. So that, for example, the average of a binary variable can be easily displayed in summary tables and graphs as 70% vs 0.7.

[ericmelse]

Option to Export a current graph to disk in the AVIF format

Although Stata supports a good number of file formats to export to, like (ps) (eps) (svg) (emf) (pdf) (png) (tif) (gif) (jpg) it does not yet support the AV1 Image File Format (AVIF).

In my opinion, the main advantage of the AVIF format is its serious reduction in file size, compared to the alternatives like jpg or png, while maintaining the same high resolution quality of the image.
This might not be much of an issue with the more mundane graphs for which a limited number of graphic objects are used.

However, the recent development of user community contributed packages to create more sophisticated visualizations of data result in graphs that have a much larger file size.
For some examples visit the Github pages of Asjad Naqvi or Ben Jann.

AVIF is an open source standard and most certainly a very serious alternative to publish graphs on web pages.
The Financial Times uses AVIF graphics to illustrate their articles, for example: When data viz goes psychotic.
Such pages load much faster because of the reduced file size, also on smart devices, which contributes to the wider digital distribution of data analysis and visualization.

[Scott Adams]

Allow variable labels longer than 80 characters.
I don't much like it either but I get survey data with variable labels such as
"14. In the past 30 days, for how many days were you TOTALLY UNABLE to carry out your usual activities or work because of any health condition?"
And while I can edit them all for brevity, maybe using a loop or something, life is short.

[Luke Masha]

If you plot a KM graph and show the at risk table under the graph, you cannot manually edit the labels of the two groups in the general fashion you can edit all other labels on a graph. The issue is the labels for the at risk table isn't selectable in the graph editor, though the numbers are. This should be fixed so that during graph edit, if the user wants to change the label of the groups in the at risk table, they can do so.

[Sergio Arancibia]

have an option to compare and find differences in 2 do files in parallel

[Abdullah Algarni]

Adopt community -coefplot- command as built-in command, with further improvement (e.g., have interactive graphic control window)

[FELICITY FORD]

@Leonardo Guizzetti i suggest the print out inform of .html format for a good preview rather than that of .txt