In this paper we introduce the Triogram method for function estimation using piecewise linear, bivariate splines based on an adaptively constructed triangulation. We illustrate the technique for bivariate regression and log-density estimation and indicate how our approach can be applied directly to model bivariate functions in the broader context of an extended linear model. The entire estimation procedure is invariant under affine transformations and is a natural approach for modeling data when the domain of the predictor variables is a polygonal region in the plane. Although our examples deal exclusively with estimating bivariate functions, the use of Triograms for modeling two-factor interactions in ANOVA decompositions of functions depending on more than two variables is straightforward. Software for fitting these models will be available in Version 4 of S.This online version of our manuscript contains a short description of Triogram modeling, and (more importantly) several of the examples from the original text. Like all of the Triogram displays in this paper, the figure at the top of the page is interactive (go ahead, position your mouse in the window, click and drag). A word of warning: we have found that hitting the reload button on your browser can really mess up an otherwise healthy Java applet.
Table of Contents
- Introduction and acknowledgements
- A short description of Triogram modeling
- A regression example with simulated data
- A bivariate density estimation example
- A regression surface with a ridge
- References
- A PostScript version of Triogram Models
- Send us comments !
KEY WORDS: Adaptive triangulations, Density estimation, Extended linear models, Finite elements, Linear splines, Multivariate splines, Regression.
Return to my home page .