Record:   Prev Next
Author Cheng, Heng-Wei
Title ARIMA models for forecasting Poisson process observations: Application to the volcanoes worldwide
book jacket
Descript 75 p
Note Source: Masters Abstracts International, Volume: 45-06, page: 3169
Adviser: Chih-Hsiang Ho
Thesis (M.S.)--University of Nevada, Las Vegas, 2007
A nonhomogeneous Poisson process (NHPP) is often appropriate for the modeling of a series of events that occur over time in a nonstationary fashion. A major difficulty with the NHPP is that it has infinitely many forms for the intensity function. In this thesis, we propose a linking bridge between a point process and the classical time series via a sequence of the empirical recurrence rates (ERR), calculated sequentially at equidistant time intervals. The distinctive technique is demonstrated with an ERR-plot, designed to fingerprint the temporal pattern of a point process. Moreover, Autoregressive Integrated Moving Average (ARIMA) models are presented to find the best fitting model to forecast the intensity associated with the underlying NHPP. Valuable modeling and computing techniques are demonstrated using volcanic data. Specifically, we split each time series data set into two groups. The first set, called the training sample, is used to develop the candidate models. The remaining data, called prediction set, is used to further evaluate the reasonableness and predictive ability of the candidate models. The information of how the candidate models considered in the model selection phase fare with the new data (prediction set), will be evaluated to conclude the "best" model
School code: 0506
Host Item Masters Abstracts International 45-06
Subject Mathematics
Alt Author University of Nevada, Las Vegas
Record:   Prev Next