2019/06/16 by Abhijit Mandal, Mandal, Abhijit
Chemistry · #62G20 #Chemistry and Stereochemistry Studies #FOS: Computer and information sciences #Methodology (stat.ME) #Primary 62G10 #Secondary 62J12
paper · pdf · doi:10.48550/arxiv.1906.06828
openalex publication_date 2019/06/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In multivariate nonparametric regression the additive models are very useful\nwhen a suitable parametric model is difficult to find. The backfitting\nalgorithm is a powerful tool to estimate the additive components. However, due\nto complexity of the estimators, the asymptotic p-value of the associated\ntest is difficult to calculate without a Monte Carlo simulation. Moreover, the\nconventional tests assume that the predictor variables are strictly continuous.\nIn this paper, a new test is introduced for the additive components with\ndiscrete or categorical predictors, where the model may contain continuous\ncovariates. This method is also applied to the semiparametric regression to\ntest the goodness-of-fit of the model. These tests are asymptotically optimal\nin terms of the rate of convergence, as they can detect a specific class of\ncontiguous alternatives at a rate of n-1/2. An extensive simulation study\nis presented to support the theoretical results derived in this paper. Finally,\nthe method is applied to a real data to model the diamond price based on its\nquality attributes and physical measurements.\n