Speaking Stata: Finding Variables

Speaking Stata: Graphing distributions

Graphing univariate distributions is central to both statistical graphics, in general, and Stata’s graphics, in particular. Now that Stata 8 is out, a review of official and user-written commands is timely. The emphasis here is on going beyond what is obviously and readily available, with pointers to minor and major trickery and various user-written commands. For plotting histogram-like display...

Speaking Stata: Problems with tables, Part I

Tables in some form or another are part and parcel of data management and analysis. The main general-purpose tabulation commands, tabulate, table, and tabstat, are reviewed and compared. When these do not provide a tabulation solution, one key strategy is to prepare the material for tabulation as a set of variables, after which the table itself can be presented with tabdisp or list. This is the...

Speaking Stata: Graphing categorical and compositional data

Abstract. A variety of graphs have been devised for categorical and compositional data, ranging from widely familiar to more unusual displays. Both official Stata commands and user-written programs are available. After a stacking trick for binary responses is explained, bar charts and related displays for cross-tabulations are discussed in detail. Tips and tricks are introduced for plotting cum...

Regression models for categorical dependent variables using Stata, 2nd Edition

Algorithms for finding dispensable variables

It is sometimes desirable to find the set of variables that have value 0 in all minimal models of a boolean formula [2]. A minimal model is one in which flipping any variable’s value from 1 to 0 leads to a non-model. All the models of the function a ⊕ b are minimal (01 and 10), and they are also the minimal models of a∨b, which has one non-minimal model (11). For both these examples the set of ...