Matias Cattaneo

Matias Cattaneo

10 packages on CRAN

binsreg

cran
99.99th

Percentile

Provides tools for statistical analysis using the binscatter methods developed by Cattaneo, Crump, Farrell and Feng (2019a) <arXiv:1902.09608> and Cattaneo, Crump, Farrell and Feng (2019b) <arXiv:1902.09615>. Binscatter provides a flexible way of describing the mean relationship between two variables based on partitioning/binning of the independent variable of interest. binsreg() implements binscatter estimation and robust (pointwise and uniform) inference of regression functions and derivatives thereof, with particular focus on constructing binned scatter plots. binsregtest() implements hypothesis testing procedures for parametric functional forms of and nonparametric shape restrictions on the regression function. binsregselect() implements data-driven procedures for selecting the number of bins for binscatter estimation. All the commands allow for covariate adjustment, smoothness restrictions and clustering.

lpdensity

cran
99.99th

Percentile

Without imposing stringent distributional assumptions or shape restrictions, nonparametric estimation has been popular in economics and other social sciences for counterfactual analysis, program evaluation, and policy recommendations. This package implements a novel density (and derivatives) estimator based on local polynomial regressions: lpdensity() to construct local polynomial based density (and derivatives) estimator, and lpbwdensity() to perform data-driven bandwidth selection.

99.99th

Percentile

Tools for statistical analysis using partitioning-based least squares regression as described in Cattaneo, Farrell and Feng (2019a, <arXiv:1804.04916>) and Cattaneo, Farrell and Feng (2019b, <arXiv:1906.00202>): lsprobust() for nonparametric point estimation of regression functions and their derivatives and for robust bias-corrected (pointwise and uniform) inference; lspkselect() for data-driven selection of the IMSE-optimal number of knots; lsprobust.plot() for regression plots with robust confidence intervals and confidence bands; lsplincom() for estimation and inference for linear combinations of regression functions from different groups.

nprobust

cran
99.99th

Percentile

Tools for data-driven statistical analysis using local polynomial regression and kernel density estimation methods as described in Calonico, Cattaneo and Farrell (2018, <doi:10.1080/01621459.2017.1285776>): lprobust() for local polynomial point estimation and robust bias-corrected inference, lpbwselect() for local polynomial bandwidth selection, kdrobust() for kernel density point estimation and robust bias-corrected inference, kdbwselect() for kernel density bandwidth selection, and nprobust.plot() for plotting results. The main methodological and numerical features of this package are described in Calonico, Cattaneo and Farrell (2019, <doi:10.18637/jss.v091.i08>).

ramchoice

cran
99.99th

Percentile

It is widely documented in psychology, economics and other disciplines that socio-economic agent may not pay full attention to all available alternatives, rendering standard revealed preference theory invalid. This package implements the estimation and inference procedures of Cattaneo, Ma, Masatlioglu and Suleymanov (2019) <arXiv:1712.03448>, which utilizes standard choice data to partially identify and estimate a decision maker's preference. For inference, several simulation-based critical values are provided.

rddensity

cran
99.99th

Percentile

Density discontinuity testing (a.k.a. manipulation testing) is commonly employed in regression discontinuity designs and other program evaluation settings to detect perfect self-selection (manipulation) around a cutoff where treatment/policy assignment changes. This package implements manipulation testing procedures using the local polynomial density estimators: rddensity() to construct test statistics and p-values given a prespecified cutoff, rdbwdensity() to perform data-driven bandwidth selection, and rdplotdensity() to construct density plots.

rdlocrand

cran
99.99th

Percentile

The regression discontinuity (RD) design is a popular quasi-experimental design for causal inference and policy evaluation. Under the local randomization approach, RD designs can be interpreted as randomized experiments inside a window around the cutoff. This package provides tools to perform randomization inference for RD designs under local randomization: rdrandinf() to perform hypothesis testing using randomization inference, rdwinselect() to select a window around the cutoff in which randomization is likely to hold, rdsensitivity() to assess the sensitivity of the results to different window lengths and null hypotheses and rdrbounds() to construct Rosenbaum bounds for sensitivity to unobserved confounders. See Cattaneo, Titiunik and Vazquez-Bare (2016) <https://rdpackages.github.io/references/Cattaneo-Titiunik-VazquezBare_2016_Stata.pdf> for further methodological details.

rdmulti

cran
99.99th

Percentile

The regression discontinuity (RD) design is a popular quasi-experimental design for causal inference and policy evaluation. The 'rdmulti' package provides tools to analyze RD designs with multiple cutoffs or scores: rdmc() estimates pooled and cutoff specific effects for multi-cutoff designs, rdmcplot() draws RD plots for multi-cutoff designs and rdms() estimates effects in cumulative cutoffs or multi-score designs. See Cattaneo, Titiunik and Vazquez-Bare (2020) <https://rdpackages.github.io/references/Cattaneo-Titiunik-VazquezBare_2020_Stata.pdf> for further methodological details.

rdpower

cran
99.99th

Percentile

The regression discontinuity (RD) design is a popular quasi-experimental design for causal inference and policy evaluation. The 'rdpower' package provides tools to perform power and sample size calculations in RD designs: rdpower() calculates the power of an RD design and rdsampsi() calculates the required sample size to achieve a desired power. See Cattaneo, Titiunik and Vazquez-Bare (2019) <https://rdpackages.github.io/references/Cattaneo-Titiunik-VazquezBare_2019_Stata.pdf> for further methodological details.

rdrobust

cran
99.99th

Percentile

Regression-discontinuity (RD) designs are quasi-experimental research designs popular in social, behavioral and natural sciences. The RD design is usually employed to study the (local) causal effect of a treatment, intervention or policy. This package provides tools for data-driven graphical and analytical statistical inference in RD designs: rdrobust() to construct local-polynomial point estimators and robust confidence intervals for average treatment effects at the cutoff in Sharp, Fuzzy and Kink RD settings, rdbwselect() to perform bandwidth selection for the different procedures implemented, and rdplot() to conduct exploratory data analysis (RD plots).