Record Details

Forecasting using decomposition and combinations of experts

DSpace at IIT Bombay

View Archive Info
 
 
Field Value
 
Title Forecasting using decomposition and combinations of experts
 
Creator GULHANE, P
MENEZES, B
REDDY, T
SHAH, K
SOMAN, SA
 
Subject artificial neural-networks
time series forecasting
neural networks
genetic algorithms
decomposition
combining techniques
 
Description We study the effect of decomposing a series into multiple components and performing forecasts on each component separately. The focus here is on sales data - most of the series considered display both seasonality and trend. Hence the original series is decomposed into trend, seasonality and an irregular component. Multiple forecasting,experts' are used to forecast each component series. These range from different feedforward neural network topologies to Holt-Winter, ARIMA (of various orders) and double exponential smoothing. We compare the forecast errors with and without decomposition. We study the result of combining using the mean/median of all expert forecasts. Since our space of composite experts runs into the thousands, we experiment with more limited cardinalities using greedy elimination and best expert pair.
 
Publisher C S R E A PRESS
 
Date 2011-10-26T14:31:48Z
2011-12-15T09:12:22Z
2011-10-26T14:31:48Z
2011-12-15T09:12:22Z
2005
 
Type Proceedings Paper
 
Identifier ICAI '05: Proceedings of the 2005 International Conference on Artificial Intelligence, Vols 1 and 2,67-73
1-932415-68-8
http://dspace.library.iitb.ac.in/xmlui/handle/10054/16036
http://hdl.handle.net/100/2632
 
Source International Conference on Artificial Intelligence (ICAI 05),Las Vegas, NV,JUN 27-30, 2005
 
Language English