Abstract:

This study analysed the Greek grain productivity values for the years 1946-2018. The achieved 73-year grain productivity values (kg to-1) have been analyzed by two mathematical methods that have been frequently used in the literature in recent years. The Algorithms of Bully Time Series (BZS) and Grey Prediction (GT) used in the field of time series are two methods of analysis used in this study. In addition to these models, a hybrid model with the BZS method has been created using the Separate Wave Conversion (ADD) technique, which is from time-series predictions methods. In the analysis phase, 53 years of grain value was used in the training of models, 20 years of grain value was used in the testing of models. The statistical success of the models has been tested by the success criteria of the error square average (HKO) and the efficiency ratio (VK). When the results were studied, the wave-blind time series (DBZS) models were found to have the highest VK value and the lowest HKO value. This method for the first time used in the field of agricultural sciences is expected to make more accurate decisions. Also, with the study, it is believed that it will keep the light to other researchers in the name of the later development of the created hybrid model.

Keywords: