COMPARING ARTIFICIAL NEURAL NETWORKS, REGRESSION, MOVING AVERAGES AND WINTERS EXPONENTIAL SMOOTHING METHODS FOR FORECASTING IN FOOD SECTOR

Abstract

In this study, a company operating in the food sector was discussed. Sales data of the company operating in the field of walnut importation between 2013-2018 years were included in the study. The purpose of the study is to examine the working problem which is explained and find results and suggestions of the study which have been discussed andto finalize the study. Consequently, with the data set created, demand estimation models which are multiple regression technique; 3, 4, and 6 moving averages techniques; singular, binary and winters' exponential correction methods and artificial neural network method are compared.