论文信息 - A Characterization of the Log-Density Smoothing Spline ANOVA Model

A Characterization of the Log-Density Smoothing Spline ANOVA Model

In this article, we introduce a characterization of the log-density smoothing spline ANOVA model. We show that in a log-density ANOVA model of order r (consisting of the main effects and all the interactions of order up to r), the joint density function is uniquely determined by the collection of all r-dimensional marginal densities. Furthermore, the order r model is the largest log-density ANOVA model under which the joint density function is uniquely determined by the r-dimensional marginals. Our results are valid for log-density ANOVA model with other general structures.

Yongho Jeon

[1] Chong Gu,et al. PENALIZED LIKELIHOOD DENSITY ESTIMATION: DIRECT CROSS-VALIDATION AND SCALABLE APPROXIMATION , 2003 .

[2] P. J. Green,et al. Density Estimation for Statistics and Data Analysis , 1987 .

[3] Graham J. Wills,et al. Introduction to graphical modelling , 1995 .

[4] E. Wegman. Nonparametric probability density estimation , 1972 .

[5] Gianluigi Pillonetto,et al. Solutions of nonlinear control and estimation problems in reproducing kernel Hilbert spaces: Existence and numerical determination , 2008, Autom..

[6] Chong Gu. Smoothing Spline Anova Models , 2002 .

[7] R. Tapia,et al. Nonparametric Probability Density Estimation , 1978 .

[8] G. Wahba. Spline models for observational data , 1990 .

[9] B. Silverman,et al. On the Estimation of a Probability Density Function by the Maximum Penalized Likelihood Method , 1982 .

[10] Yi Lin. Tensor product space ANOVA models , 2000 .

[11] Matthew P. Wand,et al. Kernel Smoothing , 1995 .

[12] Yuedong Wang,et al. Smoothing Spline Nonlinear Nonparametric Regression Models , 2004 .