Bandwidth Selection for Backfitting Estimation of Semiparametric Additive Models: A Simulation Study
2013 (English)In: Computational Statistics & Data Analysis, ISSN 0167-9473, Vol. 62, 136-148 p.Article in journal (Refereed) Published
A data-driven bandwidth selection method for backfitting estimation of semiparametric additive models, when the the parametric part is of main interest, is proposed. The proposed method is a double smoothing estimator of the mean-squared error of the backfitting estimator of the parametric terms. The performance of the proposed method is evaluated and compared with existing bandwidth selectors by means of a simulation study.
Place, publisher, year, edition, pages
2013. Vol. 62, 136-148 p.
Bandwidth selection, Semiparametric additive model, Backfitting estimation, Mean squared error, Nonparametric regression
Probability Theory and Statistics
Research subject Statistics
IdentifiersURN: urn:nbn:se:umu:diva-64072DOI: 10.1016/j.csda.2013.01.010OAI: oai:DiVA.org:umu-64072DiVA: diva2:587242