Efficient sales forecasting using PSO based adaptive ARMA model

Author

Majhi, Ritanjali ; Mishra, Sashikala ; Majhi, Babita ; Panda, Ganapati ; Rout, Minakshi

Author_Institution

Sch. of Manage., Nat. Inst. of Technol., Warangal, India

fYear

2009

fDate

9-11 Dec. 2009

Firstpage

1333

Lastpage

1337

Abstract

The paper proposes a new hybrid forecasting model using auto regressive moving average (ARMA) as basic architecture and particle swarm optimization (PSO) as learning algorithm. These two combinations have yielded an efficient prediction model for retail sales volumes. To facilitate comparison ARMA, functional link artificial neural network (FLANN) and MLP models are also simulated. The performance of the new model has been evaluated through simulation study and the results demonstrate the best prediction performance both for long and short ranges.

Keywords

autoregressive moving average processes; forecasting theory; learning (artificial intelligence); multilayer perceptrons; particle swarm optimisation; retailing; sales management; MLP models; PSO; adaptive ARMA model; auto regressive moving average; functional link artificial neural network; hybrid forecasting model; learning algorithm; particle swarm optimization; prediction model; retail sales volumes; sales forecasting; Artificial neural networks; Computer science; Computer science education; Economic forecasting; Load forecasting; Marketing and sales; Neural networks; Particle swarm optimization; Predictive models; Technology forecasting; Sales forecasting; adaptive auto regressive moving average (ARMA) model and particle swarm optimization (PSO);

fLanguage

English

Publisher

ieee

Conference_Titel

Nature & Biologically Inspired Computing, 2009. NaBIC 2009. World Congress on

Conference_Location

Coimbatore

Print_ISBN

978-1-4244-5053-4

Type

conf

DOI

10.1109/NABIC.2009.5393738

Filename

5393738