دراسة مقارنة بين نموذج الانحدار و نموذج بوكس جينكز و نماذج الشبكات العصبية لإنتاج التمور في العراق

Abstract EN

The method of artificial neural networks is one of the important new methods in building models, analysis and evaluation of data does not adopt a model or a common statistical method to diagnose the behavior of the phenomenon.

Where processing is done to reach the best model represents the phenomenon with the least possible errors and represents close to reality and can be used in most areas one of the aims of chain analysis is to build a model to explain its behavior and use the results to predict future behavior of the chain.

The aim of this research is to compare some regression models, the Box Jenkins model and the neural networks models to choose the best for the dates production variable in Iraq.

Statistical analysis software )SPSS24, minitab17, Eviews9( were used to analyze date production data for the period 1963 to 2018 and the results proved the efficiency of the neural networks method in the treatment of nonlinear models and it is considered to be a strong model.