Kristoffer Bjärkefur

SE and SD went missing when I capitalized the titles. It is fixed in 33117f7398c2bfe92caadefed6619c9fd947d8eb

Yes, `ftest` does that test. This is still accurate in the old helpfile: ``` reg testgroupdummy balancevarlist testparm balancevarlist ``` I prefer to keep it an option for backward compatibility...

I like `stats()`. And then it would be specified like `stats(pair(t) f(p))`

The options `missminmean()`, `covarmissok`, `balmiss()`, `balmissreg()`, `covmiss()`, and `covmissreg()` are among the options I said will be deprecated. They are soon gone so dont worry about them 😃

> I don't think it even needs the sub-functions -- if it always takes two arguments then it must be some combination of the specification for the estimate and the...

> "F-test" options can be added to the statistics options. No due to backward compatibility > What are missing values in F tests? I assume that this is comment in...

> Default set of comparison tests (all vs all) is good I do not understand what this means

> pairoutput option accepts multiple inputs, such as pairoutput(beta t) We decided in the meeting that this is not part of this version > causes error and returns: `option pout_lbl()...

> `pairoutput` option could be required, which would remove any issue of default functionality? No due to backward compatibility > `It could also be called something more intuitive like Stats()`...

> I am not sure that the nrmd (normalized differences) are correctly calculated. See [this link](https://blogs.worldbank.org/impactevaluations/should-we-require-balance-t-tests-baseline-observables-randomized-experiments): @bbdaniels, can you give me a correct formula? Preferable in terms of Stata code?...