Short and long-term forecasting using artificial neural networks for stock prices in Palestine: a comparative study

Authors

  • Samir K Safi The Islamic University of Gaza
  • Alexander K White Texas State University

DOI:

https://doi.org/10.1285/i20705948v10n1p14

Keywords:

Artificial Neural Network, Time Series, Forecasts, ARIMA, Regression, Stock Prices

Abstract

To compare the forecast accuracy, Artificial Neural Networks, Autoregressive Integrated Moving Average and regression models were fit with training data sets and then used to forecast prices in a test set. Three different measures of accuracy were computed: Root Mean Square Error, Mean Absolute Error and Mean Absolute Percentage Error. To determine how the accuracy depends on sample size, models were compared between daily, monthly and quarterly time series of stock closing prices from Palestine.

Author Biographies

Samir K Safi, The Islamic University of Gaza

Prof. Dr. Samir Khaled Safi is Professor of Statistics at the Islamic University of Gaza, Palestine. He earned his Master and PhD in Statistics from the American University in Washington DC. His research interests are in Time Series Analysis, Mathematical Statistics, Regression Analysis, and Econometrics. He has authored several papers concerning the efficiency of Artificial Neural Networks (ANNs) for forecasting in the presence of autocorrelated disturbances and compare ANNs with the traditional forecasting techniques. He has worked on computer simulations of time series models to determine the efficiency of ANNs in the presence of auto-correlated disturbances, he consider the robustness of various models, including Autoregressive Integrated Moving Average (ARIMA) and Regression models. He published many books, for example, "S-PLUS Programming Language and Applied Statistics", (2009) and "The Efficiency of OLS in the Presence of Auto- correlated Errors", VDM Verlag publisher (2008). In addition, Prof. Safi is the recipient of Departmental Instructorship (2000-2004), Department of Mathematics and Statistics, American University, Washington, DC, USA, Arab Student Aid International scholarship (2001-2004) and Karim Rida Said Foundation scholarship (1997-1998).

Alexander K White, Texas State University

Alexander White has a Ph.D. in Statistics from Michigan State University, is an Associate Professor of Mathematics Education and has interests in Mathematics and Statistics Education. He is currently the Advisor for the doctoral program in Mathematics Education, Vice-Chair of the Faculty Senate and Liaison to doctoral program in Developmental Education. Dr. White has worked as a consultant in test development, sample selection and evaluation for a number of organizations including the Honduran Unit for the Measurement of Educational Quality. He collaborates with Texas Mathworks, a center for mathematics education at Texas State University on curriculum development and professional development.  He recently co-authored a new middle school mathematics curriculum, Math Explorations. Dr. White has research interests in mathematics education, and statistics. In mathematics education he has authored several papers concerning the ability of students entering calculus to visualize functions and work with graphs. His has published several articles from a NSF-funded project on using Dynamic Geometry to teach high school.  He has worked on application of statistical models to assessment of educational systems in Central America.

References

Abdallah M. Al-Habeel, abdalla20022002@yahoo.com, Professor of Statistics, Al-Azhar University, Palestine.

Raed Salha, rbsalha@iugaza.edu.ps. Associate Professor of Statistics, The Islamic University of Gaz, Palestine.

Downloads

Published

26-04-2017

Issue

Vol. 10 No. 1 (2017): Electronic Journal of Applied Statistical Analysis

Section

Original Paper

License

Authors who publish with EJASA agree to the  Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia License.