.- help for ^tlogit^ Version 2.0 .- Apply the logistic tranformation to a set of variables ------------------------------------------------------ ^tlogit v1old v1new^ [v2old v2new ...]^, base(^basevar^) p^ercent Description ----------- If v1old and basevar are existing variables, ^tlogit^ will generate a new variable, v1new = ln(v1old/basevar) the log-ratio of v1old with respect to the basevar. If additional variables are specified on the left-hand side of the command, ^tlogit^ will generate additional log-ratios with respect to the basevar. Options ------- ^base(^basevar^)^ is required. It specifies the base variable for all the log- ratios. ^p^ercent indicates that the original variables (v1old, v2old...) are percentages rather than proportions. By default, tlogit assumes that the original variables are proportions, such that the values of each old variable fall between 0 and 1 inclusive and the sum of basevar and the old variables is 1. Examples -------- Suppose we have three variables -- Tory, Labour, and Ally -- which report the proportion of votes received by the Conservative party, the Labour party, and the Alliance in a particular electoral district. To create log-ratios for Tory and Labour with respect to Ally, type: . ^tlogit Tory lrTory Labour lrLabour, base(Ally)^ This command will generate two new variables, lrTory and lrLabour, the log- ratios for each party with respect to the alliance. If the original values were expressed as percentages (rather than proportions) of the total vote, we would add the ^percent^ option to create the log-ratios: . ^tlogit Tory lrTory Labour lrLabour, base(Ally) percent^ Distribution ------------ ^tlogit^ is part of CLARIFY, a suite of Stata programs for interpreting statistical results, and is (C) Copyright, 1999-2003, Michael Tomz, Jason Wittenberg and Gary King, All Rights Reserved. You may copy and distribute this program provided no charge is made and the copy is identical to the original. To request an exception, please contact: Michael Tomz Department of Political Science Encina Hall, Stanford University Stanford, CA 94305-6044 We recommend that you distribute the current version of this program, which is available from http://GKing.Harvard.Edu. References ---------- Gary King, Michael Tomz, and Jason Wittenberg (2000). "Making the Most of Statistical Analyses: Improving Interpretation and Presentation." American Journal of Political Science 44, no. 2 (April 2000): 347-61. Michael Tomz, Joshua A. Tucker, and Jason Wittenberg. 2002. "An Easy and Accurate Regression Model for Multiparty Electoral Data." Political Analysis 10, no. 1 (Winter): 66-83.