Nonparametric Estimation of Sample Selection Models

By M. Das, Whitney K. Newey, F. Vella. 2003

448 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/2483be6a5dc7005ec6262b04e2ebf174dbdefd58

How researchers see this paper

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

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

Changes in the Distribution of Male and Female Wages Accounting for Employment Composition Using Bounds (2004, unknown): “This would hold if the age effect on the qth log-wage quantile for the relevant group is additively separable and we thus call equation (15) the additivity assumption.”
Estimating distributions of potential outcomes using local instrumental variables with an application to changes in college enrollment and wage inequality (2009, methodology): “First, we estimate β1 and β0 using a semiparametric version of the sample selection estimator of Das, Newey, and Vella (2003).”
Testing exclusion restrictions and additive separability in sample selection models (2014, background): “2 As discussed in Das et al. (2003), this index restriction is implied by assuming (2) along with joint independence of (X, Z) and (U, V ) (at least conditional on Pr(S = 1|X, Z)) and a strictly monotonic cdf of V . Furthermore, Vytlacil (2002) shows that assuming a selection model with additive sep”
Trimming for Bounds on Treatment Effects with Missing Outcomes (2002, methodology): “…groups can also be seen by examining the dichotomous treatment case within the frameworks that condition on an unknown function of the probability of selection (as in Heckman and Robb 1986; Heckman 1990; Ahn and Powell 1993; Angrist 1997; Andrews and Schafgans 1998; Das, Newey, and Vella 2000).”
Counterfactual distributions with sample selection adjustments: Econometric theory and an application to the Netherlands (2007, background): “Taking the population of women working full time as given, we found an average log wage gap on the order of 15-20%, and we documented that this gap increases as we move up the distribution.”
Quantile Selection Models with an Application to Understanding Changes in Wage Inequality (2017, unknown): “[26] Das, M.”
Does patenting help high-tech start-ups? (2011, methodology): “Moreover, a Lagrange Multiplier test for ρ, capturing correlation between unobservables affecting survival and unobservables affecting firm growth, being equal to zero is rejected at the 1% level for all four specifications (1) - (4).”
Chapter 76 Large Sample Sieve Estimation of Semi-Nonparametric Models (2007, methodology): “To estimate semiparametric generalizations of Heckman's (1979) sample selection model, Gallant and Nychka (1987) proposed the Hermite polynomial sieve MLE, while Newey (1988) and Das et al. (2003) applied the series LS estimation method.”
Employer matching and 401(k) saving: Evidence from the health and retirement study (2007, methodology): “19 The estimation uses methods laid out in Vella (1992) and Das et al. (2003) for the Tobit and SCLS estimators, respectively, to correct for potential sample selection.”
Intergenerational and Intragenerational Externalities of the Perry Preschool Project (2019, unknown): “24 Our companion paper (Heckman and Karapakula, 2019) finds that the asymptotic p-values tend to overstate the 21All of the denominator probabilities for sample inclusion use a logit specification and are estimated with Greenland and Mansournia’s (2015) penalized maximum likelihood method, which avo”
Double Debiased Machine Learning Nonparametric Inference with Continuous Treatments (2019, unknown): “1, we provide sufficient low-level conditions for Assumption 3 when the nuisance estimators are kernel estimators, series, the generalized random forests in Athey, Tibshirani, and Wager (2019), and the deep neural networks in Farrell, Liang, and Misra (2021b). When there is no sample splitting (L = ”