التنبؤ لدرجات الحرارة والرطوبة النسبية لمدينة (اربيل العراق)باستخدام نماذج - بوكس جينكنز
THE EXPONENTIAL MATHEMATICAL TRANSFORMATIONS ARE USED TO IMPROVE THE QUALITY OF PREDICTIVE VALUES WITH APPLICATION ON SEASONAL DATA
In this research, the researcher took the phenomena of temperature and relative humidity, which are two series of the monthly rates of temperature and relative humidity of the city (Erbil - Iraq) for the period (2011 - 2017), and the two series of (84) value. There was a general tendency for the two temporal series (increasing trend or decreasing by months) with a few correlations and offsets of the serological values. For each of the two series, for the purpose of conversion of the two series to the state of stability around the variance was taken natural logarithm conversion, either to remove the vehicle or seasonal changes the researcher took the first seasonal difference for the seasonal period (length of season) of 12 months. Box-Jenkins models were applied to the two established series to arrive at the best seasonal model suitable for the future predictive models. We concluded that the appropriate SARIMA (3.1.4) and SARIMA (3.1) 5) 12 represent the two Celtic temperatures and relative humidity best represented using the minimum value of the Mean Square Error (MSE) criteria. The Akaike information criterion (AIC) used in the research, and after the choice of appropriate moral model and for achieving the objective of the study. The researcher uses a range of exponential mathematical transformations for obtaining the best conversion through which he can. We improve the mathematical model, which has been confirmed to the extent that it is appropriate, "The appropriate and appropriate model before and after the conversion does not change, which remains stable (Tahir, 2017). Thus, it gives predictive quality, efficiency, and use in future predictions. (Α = 0.3) and (α = 0.9) of the two series respectively (relative temperature and humidity) are considered the best exponential sports transfers to be used to improve the quality and efficiency of predictive values where the two criteria above were used mainly for differentiation.
Copyright (c) 2018 Polytechnic Journal
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.