## Access

You are not currently logged in.

Access JSTOR through your library or other institution:

Journal Article

# The $\prod$ Method for Estimating Multivariate Functions from Noisy Data

Leo Breiman
Technometrics
Vol. 33, No. 2 (May, 1991), pp. 125-143
DOI: 10.2307/1269038
Stable URL: http://www.jstor.org/stable/1269038
Page Count: 19

#### Select the topics that are inaccurate.

Cancel
Preview not available

## Abstract

The $\prod$ method for estimating an underlying smooth function of M variables, (x1,... ,xM), using noisy data is based on approximating it by a sum of products of the form $\prod_{m}\phi _{m}(x_{m})$. The problem is then reduced to estimating the univariate functions in the products. A convergent algorithm is described. The method keeps tight control on the degrees of freedom used in the fit. Many examples are given. The quality of fit given by the $\prod$ method is excellent. Usually, only a few products are enough to fit even fairly complicated functions. The coding into products of univariate functions allows a relatively understandable interpretation of the multivariate fit.

• 125
• 126
• 127
• 128
• 129
• 130
• 131
• 132
• 133
• 134
• 135
• 136
• 137
• 138
• 139
• 140
• 141
• 142
• 143