My major fields are econometric theory and applied econometrics. In particular, I am interested in identifying and estimating the preference parameters of the random utility models using multinomial discrete choice data, rank-ordered choice data, and dynamic discrete choice data. I am also interested in empirical studies that involve discrete choice data.
Publication
"Semiparametric Estimation of the Random Utility Model with Rank-Ordered Choice Data," by Jin Yan and Hong Il Yoo. Journal of Econometrics 211 (2019) 414-438.
Working paper version (2018): main text, online appendix.
Working Papers
"Semiparametric Identification and Estimation of Multinomial Discrete Choice Models using Error Symmetry," by Arthur Lewbel, Jin Yan, and Yu Zhou (under review at Quantitative Economics).
Working paper (2021): main text, online appendix.
"Trapped in Idleness: The Experience of Young Men Without Work in the United States," by Yuci Chen, Xiaoyu Xia, and Jin Yan (2021) (under review at Quantitative Economics).
This paper was presented and distributed under the title "Idle Young Men in the United States: Persistence, Educational Stepping Stones, and Longer-run Effects" before 2020.
"A Smoothed Maximum Score Estimator for Multinomial Discrete Choice Models," by Jin Yan (2017)
The previous version (2013) of the manuscript was distributed under the same title. The current version focuses on the estimation under choice-based sampling, which includes random sampling studied in the 2013 version as a special case.
"The Seemingly Unreliability of Rank-Ordered Data as a Consequence of Model Misspecification," by Jin Yan and Hong Il Yoo (2014).
Work in Progress
"Robust Estimation of Preference with Heteroskedastic and Heterogeneous Ranking Behavior," by Jin Yan (2016).
[Abstract] In this paper, I propose an estimation method to address two major concerns with the analysis of ranking data. First, the proposed method is semiparametric, thus it does not suffer from the issue of stochastic misspecification that existing parametric models suffer from. Second, it takes the unreliability of ranking behavior into consideration and provide a way to separate the consequence of unreliability of ranking data from stochastic misspecification. Heterogeneous and heteroskedastic ranking behavior is also considered in the proposed estimation method, based on which a test for the reliability of ranking information is developed.
"Semiparametric Estimation of Multinomial Discrete Choice Models with Endogenous Regressors," by Jin Yan (2013).
[Abstract] Many multinomial discrete choice applications include potentially endogenous regressors. For example, in a study of households’ choices among television options (Petrin and Train 2010), unobserved product attributes such as quality of programming are expected to be correlated with covariates such as price. To allow for endogeneity, I propose a two-stage instrumental variables estimator where the endogenous variable is replaced by a linear estimate, and then the multinomial discrete choice equation is estimated by the smoothed maximum score estimator described in my job market paper. In neither stage do I specify the distribution of the error terms, so this two-stage estimation method is semiparametric. This estimator is a generalization of the estimator proposed by Fox (2007). Fox suggests applying the maximum score estimator in the second stage of estimation. My paper is the first to derive the statistical properties of an estimator allowing for endogeneity in this semiparametric setting. The two-stage instrument variables estimator is consistent when the linear function of instrument variables and other covariates can rank order the choice probabilities.