"Stata Journal, The"

Cattaneo, Matias D.; Drukker, David M.; Holland, Ashley D.

doi: 10.1177/1536867X1301300301pmid: N/A

This article discusses the poparms command, which implements two semiparametric estimators for multivalued treatment effects discussed in Cattaneo (2010, Journal of Econometrics 155: 138–154). The first is a properly reweighted inverse-probability weighted estimator, and the second is an efficient-influence-function estimator, which can be interpreted as having the double-robust property. Our implementation jointly estimates means and quantiles of the potential-outcome distributions, allowing for multiple, discrete treatment levels. These estimators are then used to estimate a variety of multivalued treatment effects. We discuss pre- and postestimation approaches that can be used in conjunction with our main implementation. We illustrate the program and provide a simulation study assessing the finite-sample performance of the inference procedures.

Crowther, Michael J.; Hinchliffe, Sally R.; Donald, Alison; Sutton, Alex J.

doi: 10.1177/1536867X1301300302pmid: N/A

In this article, we describe a suite of commands that enable the user to estimate the probability that the conclusions of a meta-analysis will change with the inclusion of a new study, as described previously by Sutton et al. (2007, Statistics in Medicine 26: 2479–2500). Using the metasim command, we take a simulation approach to estimating the effects in future studies. The method assumes that the effect sizes of future studies are consistent with those observed previously, as represented by the current meta-analysis. Two-arm randomized controlled trials and studies of diagnostic test accuracy are considered for a variety of outcome measures. Calculations are possible under both fixed- and random-effects assumptions, and several approaches to inference, including statistical significance and limits of clinical significance, are possible. Calculations for specific sample sizes can be conducted (by using metapow). Plots, akin to traditional power curves, can be produced (by using metapowplot) to indicate the probability that a new study will change inferences for a range of sample sizes. Finally, plots of the simulation results are overlaid on extended funnel plots by using extfunnel, described in Crowther, Langan, and Sutton (2012, Stata Journal 12: 605–622), which can help to intuitively explain the results of such calculations of sample size. We hope the command will be useful to trialists who want to assess the potential impact new trials will have on the overall evidence base and to meta-analysts who want to assess the robustness of the current meta-analysis to the inclusion of future data.

Ramos-Gonñi, Juan Manuel; Rivero-Arias, Oliver; Dakin, Helen

doi: 10.1177/1536867X1301300303pmid: N/A

Reliable and accurate mapping techniques that translate health-related quality-of-life data into EQ-5D index values are now in demand by researchers conducting economic evaluation of health care technologies. In this article, we present two commands (mrs2eq and oks2eq) that translate data from two widely used disease-specific instruments into EQ-5D index values and predicted probabilities of being at a particular level on each EQ-5D domain. mrs2eq conducts a response mapping approach to transform data from the stroke-specific modified Rankin scale into index values from the generic quality-of-life EQ-5D instrument. oks2eq uses a response mapping model to estimate EQ-5D index values based on patients’ responses to the Oxford Knee Score.

Norton, Edward C.; Miller, Morgen M.; Kleinman, Lawrence C.

doi: 10.1177/1536867X1301300304pmid: N/A

In this article, we explain how to calculate adjusted risk ratios and risk differences when reporting results from logit, probit, and related nonlinear models. Building on Stata's margins command, we create a new postestimation command, adjrr, that calculates adjusted risk ratios and adjusted risk differences after running a logit or probit model with a binary, a multinomial, or an ordered outcome. adjrr reports the point estimates, delta-method standard errors, and 95% confidence intervals and can compute these for specific values of the variable of interest. It automatically adjusts for complex survey design as in the fit model. Data from the Medical Expenditure Panel Survey and the National Health and Nutrition Examination Survey are used to illustrate multiple applications of the command.

Royston, Patrick

doi: 10.1177/1536867X1301300305pmid: N/A

I provide a new tool (marginscontplot) for plotting the marginal effect of continuous covariates in regression models. The plots may be univariate or according to levels or user-selected values of a second covariate. Nonlinear relationships involving transformed covariates may be plotted on the original scale.

Riquelme, Andrés; Berkowitz, Daniel; Caner, Mehmet

doi: 10.1177/1536867X1301300306pmid: N/A

There is a growing consensus that it is difficult to pick instruments that perfectly satisfy the exclusion restriction. Drawing on results from Berkowitz, Caner, and Fang (2012, Journal of Econometrics 166: 255–266), we provide in this article a nontechnical summary of how valid inferences can be made when instrumental variables come close to satisfying the exclusion restriction. Although the Anderson–Rubin (1949, Annals of Mathematical Statistics 20: 46–63) test statistic is robust to weak identification, it assumes that the instruments are perfectly orthogonal to the structural error term and is therefore oversized under mild violations of the orthogonality condition. The fractionally resampled Anderson–Rubin (FAR) test is a modification of the Anderson–Rubin test that accounts for violations of the orthogonality condition. We show that in small samples, the size of the resampling block of the FAR test can be modified to obtain valid critical values and analyze its size and power properties. We focus on power and not on size-adjusted power because the FAR test uses only one critical value in its application. We also describe user-written commands to implement the Anderson–Rubin and FAR tests in Stata.

Hasebe, Takuya

doi: 10.1177/1536867X1301300307pmid: N/A

I describe the commands heckmancopula and switchcopula, which implement copula-based maximum-likelihood estimations of sample-selection models.

Kontopantelis, Evangelos; Reeves, David

doi: 10.1177/1536867X1301300308pmid: N/A

In this article, we describe a new individual patient data meta-analysis postestimation command, ipdforest. The command produces a forest plot following a one-stage meta-analysis with xtmixed or xtmelogit. (These commands have been renamed in Stata 13 to mixed and meqrlogit, respectively; ipdforest is currently not compatible with the new names.) The overall effect is obtained from the preceding mixed-effects regression and the study effects from linear or logistic regressions on each study, which are executed within ipdforest. Individual patient data meta-analysis models with Stata are discussed.

Blevins, Jason R.; Khan, Shakeeb

doi: 10.1177/1536867X1301300309pmid: N/A

In this article, we consider two recently proposed semiparametric estimators for distribution-free binary response models under a conditional median restriction. We show that these estimators can be implemented in Stata by using the nl command through simple modifications to the nonlinear least-squares probit criterion function. We then introduce dfbr, a new Stata command that implements these estimators, and provide several examples of its usage. Although it is straightforward to carry out the estimation with nl, the dfbr implementation uses Mata for improved performance and robustness.

Dicle, Mehmet F.

doi: 10.1177/1536867X1301300310pmid: N/A

Diversification and portfolio selection are integral parts of a finance curriculum. In this article, a multifactor capital asset pricing model is fit for components of the Dow Jones Composite Index using data from Yahoo! Finance. Along with the capital asset pricing model's Beta, other statistics that are common criteria for portfolio selection are calculated: historic standard deviation (total risk), total return, average daily return, and Sharpe and Treynor measures. Two new commands are introduced, fetchcomponents and fetchportfolio, that automate the entire process. A third new command, fetchyahoooptions, is provided to download and parse equity options data from Yahoo! Finance webpages and, optionally, to calculate the implied volatilities for the downloaded options.