NEWS (original) (raw)
descriptio 1.5
new functions
agg.wtd.mean(): aggregated weighted meanagg.wtd.var(): aggregated weighted varianceagg.wtd.quantile(): aggregated weighted quantileagg.wtd.mad(): aggregated weighted MAD
descriptio 1.4.2
bug fix
weighted.cramer(),assoc.twocat()andassoc.twocat(): bug fix when weights are not normalized (thanks to @MathieuFerry)assoc.twocat(): bug fix with weights (thanks to @MathieuFerry)
descriptio 1.4.1 [CRAN]
bug fix
regtab(): several bug fixes
descriptio 1.4 [CRAN]
new functions
crosstab(): Displays pretty 2, 3 or 4-way cross-tabulations, from possibly weighted data, and with the opportunity to color the cells of the table according to a local measure of association (phi coefficients, standardized residuals or PEM).cattab(): Bivariate statistics between a categorical variable and a set of variablescontab(): Bivariate statistics between a continuous variable and a set of variablesregtab(): Univariate and multivariate regressions and their average marginal effects side-by-sideweighted.cramer(): Cramer’s V measure of association between two (possibly weighted) categorical variablesstdres.table(): Standardized and adjusted residuals of a (possibly weighted) contingency table
minor changes
- all functions concerned : changed default NA level from “NA” to “NAs”
- all functions concerned : when na.rm = FALSE, a level for NAs is added only if there are NAs
- all functions concerned : empty levels are not automatically removed
bug fix
profiles(): bug fix when stat = “cprop” and mar = TRUE
descriptio 1.3 [CRAN]
new functions
assoc.xx(): Bivariate association measures between pairs of variablesassoc.twocat.by(): Groupwise version ofassoc.twocat()assoc.twocont.by(): Groupwise version ofassoc.twocont()assoc.catcont.by(): Groupwise version ofassoc.catcont()profiles(): Profiles by level of a categorical variable
descriptio 1.2 [CRAN]
changes in existing functions
assoc.catcont(): new items in the results (summary statistics, test-values)assoc.twocat(): new item in the results (p-values of adjusted standardized residuals)ggassoc_marimekko(): y axis labels are now horizontalcondesc()andcatdesc(): labels of the results have been renamed ; dec argument is replaced by digits and simplified ; permutation p-values can be provided for variables and categories
descriptio 1.1 [CRAN]
new functions
weighted.cor2(): weighted correlations between the columns of a data frameweighted.cov(): weighted covarianceweighted.cov2(): weighted covariances between the columns of a data frame
minor changes
ggpatternpackage moved from Imports to Suggests
descriptio 1.0 [CRAN]
Many functions in descriptio are imported fromGDAtools (1.8), with some changes and improvements, among which the main ones are :
- Every function handles weights
- Every function handles NA values
- A few function and argument name changes