The name is absent

mcross introduced by Rogers (1995) conveniently present alternative parameterizations that facilitate interpretation. While Stata
is commendably clear and accurate in explaining the meaning of the estimated parameters, in practice it is easy to be confused
about proper interpretations. For example, the zip model simultaneously estimates a binary and count model, and it is easy to
be confused on the direction of the effects.

This article describes the post-estimation command Iistcoef which allows users to list estimated coefficients in ways that
facilitate interpretation. Users can list coefficients selected by name or significance level, list transformations of the coefficients,
and request help to facilitate proper interpretation. The user can quietly estimate their model followed by the Iistcoef
command. Additional information on measures of fit for the model can be obtained with the f itstat command of Long and
Freese (2000).

Syntax

Iistcoef VyatiSst∖ [, pvalue(#) { factor ∣ percent } std constant matrix help ]

Options

pvalue(#) specifies that only coefficients significant at the # significance level or smaller will be printed. If pvalue is not
given, coefficients for all levels of significance are listed.

factor specifies that the factor change coefficients should be listed.

percent specifies that percent change coefficients should be listed instead of factor change coefficients.

std specifies that coefficients standardized to a unit variance for the independent and/or dependent variables should be listed.

For models with a latent dependent variable, the variance of the latent outcome is estimated.

constant includes the constant(s) in the output.

matrix returns results in r() class matrices. These matrices are defined below.

help includes details for interpreting each coefficient.

Description

Iistcoef can be used with the regression commands clogit, cnreg, cloglog, gologit, intreg, logistic, logit,
mlogit, nbreg, ologit, oprobit, omodel, poisson, probit, regress, zinb, and zip.

Iistcoef uses several utility ado files. These files are also used in other procedures that the authors are writing and may
be useful to other programmers. Only brief descriptions are given here. For more details, type heIp command-name after the
programs have been installed.

• _pecats.ado returns the names and values of the categories for models with ordinal, nominal, or binary outcomes. For
mlogit it indicates the value of the reference category.

• _pedum returns a scalar indicating if a variable is a dummy variable, defined as having only the values 0, 1, or missing.

• _perhs.ado returns the number of right hand side variables and their names for regression models.

• _pesum.ado computes the means, standard deviations, minima, and maxima for the variables in a regression. Optionally,
it determines the medians and whether a variable is binary. Matrices are returned with the first column containing statistics
for the dependent variables, with the remaining columns containing information for the independent variables.

Depending on the model estimated and the specified options, Iistcoef will compute standardized coefficients, factor
change in the odds or expected counts, or percent change in the odds or expected counts. The table below lists which options
and types of coefficients are available for each estimation command. If an option is the default, it does not need to be specified
in the command.

Option

Example for regress

In the simplest case, one can obtain ж-standardized, ^-standardized, and fully standardized coefficients after estimating a
model with regress. The standard Stata output is

<<   25   26   27   28   29   30   31   >>

More intriguing information