Showing papers in "Statistical Software Components in 2003"
Posted Content•
TL;DR: psmatch2 as discussed by the authors implements full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. This routine supersedes the previous 'psmatch' routine of B. Sianesi.
Abstract: psmatch2 implements full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. This routine supersedes the previous 'psmatch' routine of B. Sianesi. The April 2012 revision of pstest changes the syntax of that command.
1,887 citations
Posted Content•
TL;DR: In this paper, a finite-sample correction to the two-step covariance matrix derived by Windmeijer et al. is proposed to make twostep robust more efficient than onestep robust.
Abstract: xtabond2 can fit two closely related dynamic panel data models. The first is the Arellano-Bond (1991) estimator, which is also available with xtabond without the two-step finite-sample correction described below. The second is an augmented version outlined in Arellano and Bover (1995) and fully developed in Blundell and Bond (1998). Arellano and Bond (1991) developed a Generalized Method of Moments estimator that treats the model as a system of equations, one for each time period. The equations differ only in their instrument/moment condition sets. The predetermined and endogenous variables in first differences are instrumented with suitable lags of their own levels. Strictly exogenous regressors, as well as any other instruments, can enter the instrument matrix in the conventional instrumental variables fashion: in first differences, with one column per instrument. A problem with the original Arellano-Bond estimator is that lagged levels are often poor instruments for first differences, especially for variables that are close to a random walk. Arellano and Bover (1995) described how, if the original equations in levels were added to the system, additional moment conditions could be brought to bear to increase efficiency. In these equations, predetermined and endogenous variables in levels are instrumented with suitable lags of their own first differences. Blundell and Bond (1998) articulated the necessary assumptions for this augmented estimator more precisely and tested it with Monte Carlo simulations. The original estimator is sometimes called "difference GMM," and the augmented one, "system GMM." Bond (2002) is a good introduction to these estimators and their use. xtabond2 implements both estimators. As GMM estimators, the Arellano-Bond estimators have one- and two-step variants. But though asymptotically more efficient, the two-step estimates of the standard errors tend to be severely downward biased (Arellano and Bond 1991; Blundell and Bond 1998). To compensate, xtabond2, unlike xtabond, makes available a finite-sample correction to the two-step covariance matrix derived by Windmeijer (2000). This can make twostep robust more efficient than onestep robust, especially for system GMM. Note: the routine requires an up-to-date version of Stata 7 (with the 21jun2002 update installed). Users of Stata version 10 (25feb2008 update) or later can take advantage of speed improvements due to Mata.
236 citations
Posted Content•
TL;DR: These graphical displays are used to examine whether the results of a meta-analysis may have been affected by publication or other types of bias.
Abstract: metafunnel plots funnel plots. These graphical displays are used to examine whether the results of a meta-analysis may have been affected by publication or other types of bias.
14 citations
Posted Content•
TL;DR: Inverse Inverse is used for calculating geodesic distances between a pair of points on the surface of the Earth (specified in signed decimal degrees latitude and longitude) using an accurate ellipsoidal model of Earth.
Abstract: vincenty is used for calculating geodesic distances between a pair of points on the surface of the Earth (specified in signed decimal degrees latitude and longitude), using an accurate ellipsoidal model of the Earth. see http://www.ngs.noaa.gov/PUBS_LIB/inverse.pdf
13 citations
Posted Content•
TL;DR: mvprobit as mentioned in this paper uses the Geweke-Hajivassiliou-Keane (GHK) simulator to evaluate the M-dimensional Normal integrals in the likelihood function.
Abstract: mvprobit estimates M-equation probit models, by the method of simulated maximum likelihood (SML). (Cf. probit and biprobit which estimate 1-equation and 2-equation probit models by maximum likelihood.) The variance-covariance matrix of the cross-equation error terms has values of 1 on the leading diagonal, and the off-diagonal elements are correlations to be estimated. mvprobit uses the Geweke-Hajivassiliou-Keane (GHK) simulator to evaluate the M-dimensional Normal integrals in the likelihood function. For each observation, a likelihood contribution is calculated for each replication, and the simulated likelihood contribution is the average of the values derived from all the replications. The simulated likelihood function for the sample as a whole is then maximized using standard methods (ml in this case).
9 citations
Posted Content•
TL;DR: factortest performs Bartlett's test for sphericity and calculates the Kaiser-Meyer-Olkin Measure of Sampling Adequacy as mentioned in this paper, both tests should be used prior to a factor or a principal component analysis.
Abstract: factortest performs Bartlett's test for sphericity and calculates the Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Both tests should be used prior to a factor or a principal component analysis.
8 citations
Posted Content•
TL;DR: In this article, the authors present a two-parameter beta distribution for Stata 4 through 7, with a single variable dependent on covariates, which can be used in Stata 8.
Abstract: betafit fits a two-parameter beta distribution, optionally as dependent on covariates. Stata 8 is required. Also included in this package is beta4, which also fits a two-parameter beta distribution, although to a single variable only. That may be used in Stata 4 through 7. This routine was previously circulated as beta.
6 citations
Posted Content•
TL;DR: The hireg command conducts hierarchical regressions which enter blocks of independent variables which are added to the model in successive steps and report R-squared change at each step.
Abstract: The hireg command conducts hierarchical regressions. Users enter blocks of independent variables which are added to the model in successive steps. R-squared change is reported at each step along with a summary table at the end. All options available on the regress command may also be used with hireg.
5 citations
Posted Content•
TL;DR: Stripplot plots data as a series of marks against a single magnitude axis, horizontal or vertical as an Stata 8 alternative to gr7, oneway or dotplot with more flexibility in several respects.
Abstract: stripplot plots data as a series of marks against a single magnitude axis, horizontal or vertical. It is an Stata 8 alternative to gr7, oneway or dotplot with more flexibility in several respects. Its predecessor, onewplot, remains available to users of Stata 7 or Stata 6 as a separate module.
2 citations
Posted Content•
TL;DR: In this paper, the authors present a method to estimate the probability of occurrence of each event of interest and one or more competing events whose occurrence precludes or alters the probabilities of occurence of the first one.
Abstract: In survival or cohort studies the failure of an individual may be one of several distinct failure types. In such a situation we observe an event of interest and one or more competing events whose occurrence precludes or alters the probability of occurence of the first one. stcompet creates variables containing Cumulative Incidence, a function that in this case appropriately estimates the probability of occurrence of each endpoint, corresponding Standard Error and Confidence Bounds. The values in numlist of the previous stset are assumed as occurrence of event of interest. In compet() options you can specify numlist relating to the occurrence of up to six competing events. This version has been updated from that published in Stata Journal, 4:2.
2 citations
Posted Content•
TL;DR: eclplot as discussed by the authors creates a plot of estimates with lower and upper confidence limits on one axis against another numeric variable on the other axis, in a data set with one observation per confidence interval to be plotted.
Abstract: eclplot creates a plot of estimates with lower and upper confidence limits on one axis against another numeric variable on the other axis. The estimates and lower and upper confidence limits are stored in three variables, in a data set with one observation per confidence interval to be plotted. Data sets with such variables may be created manually (using a spreadsheet), or using the parmest package, or using statsby or postfile in official Stata. The user has a choice of plotting the confidence intervals horizontally or vertically, a choice of plot styles for the estimates, and a choice of range plot styles for the confidence intervals. It is also possible to overlay the confidence interval plot with other plots using the plot() option, and to overlay multiple superimposed confidence interval plots using the supby() option. eclplot may be used either as a command or through a dialog.
Posted Content•
TL;DR: pcorr2 as mentioned in this paper is a modified and enhanced version of the pcorr command, which only displays partial correlations and can be used to display partial and semipartial correlation coefficients.
Abstract: pcorr2 displays partial and semipartial correlation coefficients. It is a modified and enhanced version of the pcorr command, which only displays partial correlations. The syntax and other abilities of pcorr and pcorr2 are identical.
Posted Content•
TL;DR: The bigtab command helps when you get a "too many values" error from tabulate or want to save the results in a data file.
Abstract: The bigtab command helps when you get a "too many values" error from tabulate or want to save the results in a data file Users may specify up to three variables and have the option of displaying and saving cumulative, row, and column counts and percentages