R2sls package features

From InterSciWiki
Jump to: navigation, search

Eff and Dow 2009 2SLS (2-stage least squares) prototype with improved features

  1. Missing data imputation using completely coded auxiliary data
  2. Improved by choice of full or conditional imputation (for the subsample coded on the dependent variable)
  3. Improved by additional relative effect reff coefficient in addition of significance that acts like a do measure (analogous to percentages)
  4. Improved by inferential statistics derived from random samples of observations to estimate model coefficients then used to rest replication of significance using the remaining independent subsamples.
  5. Significance tests efficiently estimated in 2SLS
  6. First stage OLS autocorrelation controls with Instrumental variables (Distance, Language)
  7. Second stage OLS measuring randomness of error terms using Instrumental variables
  8. Diagnostics for goodness-of-fit
  9. Abstract
  10. full article

White and White 2011 R2sls-plus features

  1. Refactoring of code as a functional R package
  2. Calculation of effect ratios following Pearl (2010)
  3. Network diagrams
  4. Identification of dependencies and adjustments of causal effects
  5. Split halves test
  6. Probit analysis (transforming ordinal to interval variables)
  7. Data quality controls, measurement error and unmeasured confounds
  8. Structural equation modeling (SEM) with a causal component (Kyono 2010) guided by Pearl's d-seperation, prior to data analysis.
  9. Direct and indirect effects, example: Polygyny Model

Software

Abstract: Structural equation modeling (SEM) is the leading method of causal inference in the behavioral and social sciences and, although SEM was first designed with causality in mind, the causal component has been obscured and lost over time due to a lack of adequate formalization. The objective of this thesis is to introduce and integrate recent advancements in graphical models into SEM through user-friendly software modules, in hopes of providing SEM researchers valuable information, extracted from path diagrams, to guide analysis prior to obtaining data. This thesis presents a software package called Commentator that assists users of EQS, a leading SEM tool. The primary function of Commentator is to take a path diagram as input, perform analysis on the graphical input, and provide users with relevant causal and statistical information that can subsequently be used once data is gathered. The methods used in Commentator are based on the d-separation criterion, which enables Commentator to detect and list: (i) identifiable parameters, (ii) identifiable total effects, (iii) instrumental variables, (iv) minimal sets of covariates necessary for estimating causal effects, and (v) statistical tests to ensure the compatibility of the model with the data. These lists assist SEM practitioners in deciding what test to run, what variables to measure, what claims can be derived from the data, and how to modify models that fail the tests.
  • Appendix 12: pp 33-56. Code compiled in Microsoft Visual Studio 2008. (suitable for recoding in R).

Download