Bias reduction is an important condition for effective feature extraction. Utilizing recent theoretical results in high dimensional statistical modeling, we propose a model-free yet computationally simple approach to estimate the partially linear model Y = Xβ+g(Z)+ε. Based on partitioning the support of Z, a simple local average is used to approximate the response surface g(Z). The model can be...

A semi-parametric mixed eeects regression model is proposed for the analysis of clustered or longitudinal data with continuous, ordinal or binary outcome. The commonly used Gaussian assumption on the random eeects distribution is relaxed as the partial predictive recursion method (Newton and Zhang, 1996) is used to estimate the random eeects density function. The parameter estimates are obtaine...

Non-parametric graphical models, embedded in reproducing kernel Hilbert spaces, provide a framework to model multi-modal and arbitrary multi-variate distributions, which are essential when modeling complex protein structures. Non-parametric belief propagation requires the structure of the graphical model to be known a priori. Currently there are nonparametric structure learning algorithms avail...

Mixed effects models are frequently used for analyzing longitudinal data. Normality assumption of random effects distrbution is a routine assumption for these models, violation of which leads to model misspecification and misleading parameter estimates. We propose a semi-parametric approach using gradient function for random effect estimation. In the approach, we relax the normality assumption ...

A new approach to short-term electrical load forecasting is investigated in this paper. As electrical load data are highly non-linear in nature, in the proposed approach, we first separate out the linear and the non-linear parts, and then forecast using the non-linear part only. Semi-parametric spectral estimation method is used to decompose a load data signal into a harmonic linear signal mode...

This study extends the parametric estimation of the structural misclassification model (Arrieta, 2008) to analyze over-treatment of medical procedures to a semi-parametric estimation. The estimation is based on a doubleindex semi-parametric maximum likelihood with partial observability. The document shows that misspecification error due to an incorrect assumption about the error distribution ma...