模型及其估计与R

1. Estimation method

standard error estimation

sanwich, 可以通过AER::coeftest(), modelsummary::modelsummary()来调用vcov,也可以在一些模型中直接调用，比如fixest::feols().

Simulation-Based Inference

clarify, Simulation-Based Inference for Regression Models

bootstrap

In R, you can use the cluster option in the vcovBS() function in the sandwich package if what you want is bootstrap standard errors on a regression model (pass the vcovBS() result to the vcov argument in your modelsummary table).

Wild bootstrap standard errors can be implemented in R using the type argument of vcovBS() in the sandwich package, or there are some faster options in the fwildclusterboot package.

boot: Bootstrap Functions (Originally by Angelo Canty for S)

momentfit: Methods of Moments

momentfit:Several classes for moment-based models are defined. The classes are defined for moment conditions derived from a single equation or a system of equations.

2. Models

  • Weighting

WeightIt, 通过加权的方法调整变量之间的差异。

  • Matching

MatchIt, 种匹配方法的集合。

causalweight: Estimation Methods for Causal Inference Based on Inverse Probability Weighting

  • Fixed effect

fixest, Fast and user-friendly fixed-effects estimation

  • event study

estudy2, An implementation of a most commonly used event study methodology, including both parametric and nonparametric tests.

  • difference in difference

did, The did package contains tools for computing average treatment effect parameters in a Difference-in-Differences setup allowing for More than two time periods; Variation in treatment timing (i.e., units can become treated at different points in time); Treatment effect heterogeneity (i.e, the effect of participating in the treatment can vary across units and exhibit potentially complex dynamics, selection into treatment, or time effects); The parallel trends assumption holds only after conditioning on covariates.

The DRDID R package implements different estimators for the Average Treatment Effect on the Treated (ATT) in Difference-in-Differences (DID) setups where the parallel trends assumption holds after conditioning on a vector of pre-treatment covariates.

  • the sunab function in the fixest package

  • etwfe, an package implementing extended TWFE proposed by Wooldridge.

  • synthetic control

Synth: Synthetic Control Group Method for Comparative Case Studies

scpi, implementations of estimation and inference procedures for synthetic control methods.

  • instrumental variable

AER::ivreg() function in the AER package

ivmodel: Statistical Inference and Sensitivity Analysis for Instrumental Variables Model

ivtools: Instrumental Variables：Contains tools for instrumental variables estimation.

  • Regression discontinuity Designs

RD series package: Software packages for analysis and interpretation of regression discontinuity designs and related methods. Replication files and illustration codes employing these packages are also available.

RDHonest，This R package implements honest and efficient confidence intervals in fuzzy and sharp regression discontinuity designs using procedures from Armstrong and Kolesár (2020) (preprint), Armstrong and Kolesár (2018) (preprint), and Kolesár and Rothe (2018) (preprint).

3. Interpretation

  • marginaleffects, Compute and plot predictions, slopes, marginal means, and comparisons (contrasts, risk ratios, odds ratios, etc.) for over 70 classes of statistical models in R. Conduct linear and non-linear hypothesis tests, as well as equivalence tests using the delta method.

  • effectsize, The goal of this package is to provide utilities to work with indices of effect size and standardized parameters, allowing computation and conversion of indices such as Cohen’s d, r, odds-ratios, etc

  • ggeffects, ggeffects: Tidy Data Frames of Marginal Effects from Regression Models.

  • modelbased, a package helping with model-based estimations, to easily compute of marginal means, contrast analysis and model predictions.