Publications Search

Abstract

Title: A Mixture Model for the Probability Distribution of Rain Rate.
Author: Jeffries N, Pfeiffer R
Journal: Environmetrics 12:1-10
Year: 2000
Month: None

Abstract: In this paper we present a logistic mixture model for rain rate, that is, a model where the regime probabilities are allowed to change over time and are modeled with a logistic regression structure. Such a model may be used as an alternative to simple mixture, threshold, or hidden Markov models. The maximum likelihood estimates for the model parameters are found using an EM algorithm and their asymptotic properties are stated. The model is fit to hourly measurements of rain rate that are part of the GATE dataset. The results are compared with results from a standard mixture model and from a single density model. Copyright © 2001 John Wiley & Sons, Ltd.