Inverse probability weighted estimation for general missing data problems

By Jeffrey M. Wooldridge. 2004

1,011 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Recommendations as Treatments: Debiasing Learning and Evaluation (2016, background): “This is a known effect, partly because the estimated propensities can provide an effect akin to stratiﬁcation (Hi-rano et al., 2003; Wooldridge, 2007).”
The Small Firm Effect and the Entrepreneurial Spawning of Scientists and Engineers (2010, methodology): “20 To account for the potential nonrandom selection into small firms implied by explanation (b), we reestimate the performance Equation (3) adopting the inverse propensity score weighting methods discussed in Hirano and Imbens (2001) and Wooldridge (2007).21 These methods allow us to more fully capt”
A New Instrumental Method for Dealing with Endogenous Selection (2010, methodology): “This estimator enables in turn to make valid inference on the whole population, by an inverse probability weighting procedure, in a similar fashion to [20], [24], [40] or [45].”
Doubly robust uniform confidence band for the conditional average treatment effect function (2016, methodology): “nce then, many other alternative doubly robust estimators have been proposed in the literature. For example, the inverse probability weighting regression adjustment estimator (Kang and Schafer, 2007; Wooldridge, 2007, 2010) is widely known and has been implemented in statistical software packages. S”
Semiparametric Estimation and Inference Using Doubly Robust Moment Conditions (2013, neutral): “See for example Robins, Rotnitzky, and Zhao (1994), Robins and Rotnitzky (1995), Scharfstein, Rotnitzky, and Robins (1999), Robins and Rotnitzky (2001), Van der Laan and Robins (2003) or Tan (2006), and Wooldridge (2007) or Graham, Pinto, and Egel (2012) for applications in econometrics.”
Tax me, but spend wisely? Sources of public finance and government accountability (2015, background): “Hirano et al. (2003) show that this estimator is efficient and Wooldridge (2007) that ignoring the first-stage estimation of the selection probabilities when performing inference yields conservative standard errors.”
Estimation of Conditional Average Treatment Effects With High-Dimensional Data (2019, background): “(38) (Although Belloni and Chernozhukov (2011) focus on the high-dimensional quantile regression without approximation error, it is straightforward to extend their results to the case with approximation error as long as it is sufficiently small, as imposed in Assumption 4.”
Semiparametric efficiency in GMM models with auxiliary data (2007, positive): “…in models with attrition or nonresponse [e.g., Little and Rubin (2002), Robins and Rotnitzky (1995), Rotnitzky and Robins (1995), Wooldridge (2002), Wooldridge (2003)], the estimation of treatment effects [see e.g., the references surveyed in Heckman, LaLonde and Smith (1999)], the recovery of…”
Impacts of extension access and cooperative membership on technology adoption and household welfare (2017, methodology): “…of conditional independence and the overlap conditions, ATT is computed as follows: ATT ¼ E½Yð1ÞjT ¼ 1; pðxÞ E½Yð0ÞjT ¼ 0;pðxÞ (3) However, ATT from PSM can still produce biased results in the presence ofmis-specification in the propensity scoremodel (Robins et al., 2007; Wooldridge, 2007, 2010).”
Doubly robust difference-in-differences estimators (2018, neutral): “Assumption 4 is standard in the literature, see e.g. Abadie (2005), Wooldridge (2007), Bonhomme and Sauder (2011) and Graham et al. (2012).”
Inverse Probability Tilting for Moment Condition Models with Missing Data (2012, neutral): “Wooldridge (2007) makes a related point: equation (4) cannot increase if the dimension of r increases.”
Estimating average partial effects under conditional moment independence assumptions (2004, result): “It follows from the general results in Wooldridge (2003b) that ignoring the estimated propensity score, using (4.5), (4.6), or (4.7), results in conservative standard errors.”