dm0042_4: Speaking Stata: Distinct observations. N. J. Cox and G. M. Longton. Stata Journal 20: 1028–1030; Stata Journal 15: 899; Stata Journal 12: 352; Stata Journal 8: 557–568.
The most important change in this update is the addition of the distinctgen command, which generates a variable with the number(s) of distinct observations or values. Commonly, this will be used with a by() option to generate numbers of distinct observations or values within different groups of observations.
A fix to distinct ensures that zeros are returned as stored results whenever no observations satisfy the specified conditions.
dm0085_3: Speaking Stata: A set of utilities for managing missing values. N. J. Cox. Stata Journal 20: 1028; Stata Journal 17: 779; Stata Journal 15: 1174–1185.
The most important change in this update is the addition of the subcommand breakdown, which issues a report on different missing values; that is, the numbers present as 1) empty strings "" if string variables are included and 2) system missing and extended missing values if numeric variables are included. This subcommand is most obviously useful as a check on the presence of extended missing values for numeric variables.
st0175_3: A menu-driven facility for sample-size calculation in novel multiarm, multistage randomized controlled trials with a time-to-event outcome. B. Choodari- Oskooei, P. Royston, I. R. White, and M. K. B. Parmar. Stata Journal 19: 782–802 (A. Blenkinsop and B. Choodari-Oskooei); Stata Journal 15: 350–368 (D. J. Bratton, B. Choodari-Oskooei, and P. Royston); Stata Journal 9: 505–523 (F. M.-S. Barthel, P. Royston, and M. K. B. Parmar).
The updated nstage command calculates the sample size for multiarm, multistage (MAMS) designs under the alternative variance formula for the log hazardratio and adds an option to calculate the sample size under the null variance formula. (Sample sizes derived using the previous version of nstage can be calculated using the old option.) We have also produced video tutorials on nstage (see https://www.mrcctu.ucl.ac.uk/our-research/methodology/design/more-aboutmams/mams-workshop-videos-nstage-examples/) that can help new users and can help facilitate the implementation of MAMS designs in different trial settings. A note describing and illustrating the changes is available from the following link: https://www.mrcctu.ucl.ac.uk/media/2326/nstage-update-main-article-final.pdf.
st0284_1: Trial sequential boundaries for cumulative meta-analyses. B. Miladinovic, I. Hozo, and B. Djulbegovic. Stata Journal 13: 77–91.
The metacumbounds command has been updated to reflect the 2022 updates to the ldbounds package in R. This change does not affect previous results obtained; it adds only improved functionality to the command.
st0606_1: Comparing distributions of ordinal data. S. P. Jenkins. Stata Journal 20: 505–531.
A minor bug has been squashed: the returned scalar for the estimate of the Cowell–Flachaire index, I(0.25), now contains the correct value rather than the value for I(0.75).
