B
Brian D. Marx
Researcher at Louisiana State University
Publications - 89
Citations - 6832
Brian D. Marx is an academic researcher from Louisiana State University. The author has contributed to research in topics: Generalized linear model & Regression analysis. The author has an hindex of 26, co-authored 87 publications receiving 6120 citations.
Papers
More filters
Journal ArticleDOI
Flexible smoothing with B-splines and penalties
Paul H. C. Eilers,Brian D. Marx +1 more
TL;DR: A relatively large number of knots and a difference penalty on coefficients of adjacent B-splines are proposed to use and connections to the familiar spline penalty on the integral of the squared second derivative are shown.
Book
Regression: Models, Methods and Applications
TL;DR: The Classical Linear Model is extended to include nonparametric Regression and Structured Additive Regression in the model of Quantile Regression.
Journal ArticleDOI
Direct generalized additive modeling with penalized likelihood
Brian D. Marx,Paul H. C. Eilers +1 more
TL;DR: A difference penalty on adjacent B-spline coefficients is incorporated into a penalized version of the Fisher scoring algorithm, and each component has a separate smoothing parameter, and the penalty is optimally regulated through extensions of cross validation or information criterion.
Journal ArticleDOI
Splines, knots, and penalties
Paul H. C. Eilers,Brian D. Marx +1 more
TL;DR: In this article, the authors compare B-splines with difference penalties with truncated power functions, knots based on quantiles of the independent variable and a ridge penalty, and conclude that the difference penalties are clearly to be preferred.
Journal ArticleDOI
Generalized linear regression on sampled signals and curves: a P -spline approach
Brian D. Marx,Paul H. C. Eilers +1 more
TL;DR: In this paper, the authors consider generalized linear regression with many highly correlated regressors, which requires severe regularization because the number of regressors is large, often exceeding the total number of observations, and solve the problem by forcing the coefficient vector to be smooth on the same domain.