Estimation of Quantile and other Nonparametric Statistical Functionals in Non- and Semiparametric Regression and Autoregression Models

• Applicant: Professor Dr. Enno Mammen
• Subject Area: Mathematics
• Term: from 2006 to 2011
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 21235440

• Final Report Year: 2011

Final Report Abstract

1. Higher dimensional nonparametric models with several nonparametric components have been analysed in different dependence structures, including nonstationarity and local stationarity. 2. New GARCH-models have been developed that include nonparametric components in their dynamics. 3. In the setting of density estimation estimation and visualization methods were developed to cope with high dimensional cases. An asymptotic theory has been developed for their performance. 4. Higher dimensional nonparametric quantile models have been studied and an asymptotic theory for backfitting estimators has been developed.

Publications

• (2009). L2 boosting in kernel regression. Bernoulli 15, 599-613
  B.U. Park, Y. K. Lee and S. Ha
  
• (2009). Multivariate histograms with data-dependent partitions. Statistica Sinica 19, 159-176
  Klemelä, J.
  
• (2009). Smoothing of Multivariate Data: Density Estimation and Visualization. Wiley, New York
  Klemelä, J.
  
• (2010). Nonparametric Estimation of Distributional Policy Effects. Journal of Econometrics 155, 56-70
  Rothe, C.