|
|
Melvin D'Vaz
dvazm01@student.uwa.edu.au
Entry year: 2008
Enrolment status: checked and confirmed
Degree: BE(SE)
Degree status: continuing
Project: Rare Event Tracker Algorithm
Supervisor(s): Tim French, Mark Reynolds, and David Coward (Physics)
Project status: expects to complete end of semester 2, 2008
|
|
This project involves modifying and analysing an existing algorithm, which is currently used for forecasting certain types of sporadic events.
Time series forecasting has been around for a long time.
The general theory is that all time series contain a random component, which must be filtered out, and then a prediction can be made.
The rare event tracker algorithm, developed by Howell and Coward, has no need to filter out this random component, and can fit curves to different types of time series because of this.
However, this algorithm only works on time series that follow a Poisson distribution.
This project will generalizing this algorithm to remove its dependency on Poisson distributions, hence being able to apply it in a wider range of context.
Its effectiveness as a forecasting tool will then be tested.
|
 Proposal
|
Last update: Thu Mar 27 15:49:33 2008
|
|
