Comparative Study of ARIMA & ANN Models - Saha, Shishir;Das, Sumonkanti;Chowdhury, Md. Ahmed Kabir
32,99 €
inkl. MwSt.
Versandkostenfrei*
Versandfertig in 6-10 Tagen
  • Broschiertes Buch

Time series analyses particularly forecasting of financial data have been attracting some special attention due to its nonlinear nature and complex behavior. Since the classical statistical methods become inadequate in such cases, a new generation of methodologies has been using for the analysis of trends, patterns and forecasting. Artificial Neural Networks (ANN) has been approved successful in recent time to approximate such non-linearity. In pursuing the prediction of Dhaka Stock Exchange (DSE) General Index, the study has made an attempt to develop a traditional linear ARIMA model and a…mehr

Andere Kunden interessierten sich auch für
Produktbeschreibung
Time series analyses particularly forecasting of financial data have been attracting some special attention due to its nonlinear nature and complex behavior. Since the classical statistical methods become inadequate in such cases, a new generation of methodologies has been using for the analysis of trends, patterns and forecasting. Artificial Neural Networks (ANN) has been approved successful in recent time to approximate such non-linearity. In pursuing the prediction of Dhaka Stock Exchange (DSE) General Index, the study has made an attempt to develop a traditional linear ARIMA model and a non-linear ANN model. Finally, the prediction performances are compared on the basis of accuracy in prediction and several performance indicators. The findings of the study suggest ANN model as the more efficient one for the DSE General Index prediction than the traditional ARIMA model.
Autorenporträt
S Saha obtained both BSc and MS degrees in Statistics from Shahjalal University of Science & Technolgy, Sylhet, Bangladesh. He did his MS Thesis on Time Series Modeling of DSE share price index under supervision of Professor MAK Chowdhury and Mr. S Das. Saha is interested in Market Research, Monitoring & Evaluation, and Data Management.