`balanceTest()` examples for less common answers, function arguments

[benthestatistician]

To be addressed in balanceTest() examples:

• Non-NULL post.alignment.transform= argument
• Non-default stratum.weights= argument
• Return values other than 1.0 for "non-null record" means/imbalances

Notes

Non-NULL post.alignment.transform

As of this writing, there's a minor problem with the current example and a larger problem w/ the argument itself. (The larger problem being beyond the current issue's scope, noted here to track possibility of its being addressed elsewhere, changing or obviating the minor problem.)

Minor problem: in the example below, the unstratified case gives Wilcoxon tests, as stated, but the stratified case gives a bundle of aligned rank tests.

##' ##Comparing unstratified to stratified, just one-by-one wilcoxon
##' #rank sum tests and omnibus test of multivariate differences on
##' #rank scale.
##' balanceTest(pr~ date + t1 + t2 + cap + ne + ct + bw + cum.n + strata(pt),
##'          data=nuclearplants,
##'          report=c("adj.means", "chisquare.test"),
##'      post.alignment.transform=rank)

(The larger problem is that in the case with clusters and/or unit weights, the post.alignment transform is being applied after aggregation, whereas it might have been more natural to apply it before aggregation [#64].)

Non-default stratum.weights=

Issue #100 was resolved [master d086a01] in such a way that users should be able to control stratum weights used both for descriptive and inferential statistics. We should check whether this is the case, and if so provide an example and perhaps a test (of internal consistency between inferential and descriptive calculations that are returned).

Return values other than 1.0 for "non-null record" means/imbalances

An example is needed to show what this means, if/when it arises. Except that I'm not sure it's currently every being reported; see comment in #85.