Stefan Sperlich’s research published in three top-tier journals

GSEM Professor Stefan Sperlich’s research in the fields of statistics and econometrics has been accepted for publication in three top-tier journals: Journal of the American Statistical Association, Biometrika, and Journal of Business & Economic Statistics.

In a paper published in the Journal of the American Statistical Association, the authors develop simultaneous confidence intervals and multiple testing procedures for empirical best predictors under generalized linear mixed models. A case study on predicting poverty rates illustrates the applicability and advantages of their simultaneous inference tools. The article was co-authored by Katarzyna Reluga (University of Cambridge) and María-José Lombardía (University of A Coruña). This article builds on Dr. Reluga’s thesis, which she obtained from the GSEM in 2020 under the co-guidance of Professors Sperlich and Lombardía.

The article published in Biometrika introduces backfitting tests for semiparametric generalized structured models, which can be used for testing different kinds of model specifications, or for performing variable selection in a large class of semiparametric models. The paper was co-authored with Enno Mammen (Heidelberg University).

Finally, the article published in the Journal of Business & Economic Statistics presents an adaptive omnibus specification test of asset pricing models where the stochastic discount factor is conditionally affine in the pricing factors. The test is robust to functional form misspecification and detects any relationship between pricing errors and conditioning variables. The paper was co-authored by Francisco Peñaranda (Queens College CUNY) and Juan M. Rodríguez-Poo (University of Cantabria).

Stefan Sperlich joined the GSEM in 2010 as a Professor of Statistics and Econometrics. He holds a Ph.D. in Economics from the Humboldt University of Berlin, and his research interests range from nonparametric statistics over small area statistics to empirical economics, in particular impact evaluation methods.