PENERAPAN METODE POLINOM NEWTON GREGORY MAJU DAN POLINOM NEWTON GREGORY MUNDUR DENGAN METODE HAMILTON-PERRY DALAM MEMPREDIKSI JUMLAH PENDUDUK SUMATERA UTARA.

Abstract

Population data needs are absolute things that must be met by BPS, but the most complete source of population data comes from population census results. Meanwhile, the implementation of the population census is only once every 10 years because it requires cost, time, and energy. Population prediction in the following year in each census period needs to be done to determine the difference in population growth in that year. Therefore, researchers are interested in studying the "Application of the Advanced Newton Gregory Polynomic Method and the Newton Gregory Backward Polynomic with the Hamilto-Perry Method in Predicting the Population of North Sumatra." this case the researcher will analyze between the three census periods. The purpose of this study is to obtain effective results from the three methods so that it can be used as an alternative application for the Central Statistics Agency (BPS) of North Sumatra province to predict the population in the year between the census. For Hamilton-perry, the application of the method is easy to do because the calculation process is uncomplicated and the data used as prediction material is very minimal, but the predicted value produces a non-rounded value so it needs to be rounded but when the value has been rounded there is a difference between the value not rounded to the rounded value aggressively. Therefore we need an adjustment to the value of rounding results based on the value not rounded. Then based on an aggregate trial of population prediction using the Hamilton-Perry method it produces a value that is higher than the actual value. For the advanced newton gregory polynomic method and the backward newton gregory polynomic each obtained total relative error isΣ𝜀𝑅=0,31412327 and Σ𝜀𝑅=0,3522818. From the two relative errors it is known that the total error for the advanced Newtonton polynomial method is smaller than the total error for the backward Newtonian polynomial method. Thus the Newton Newton method of advanced polynomial has better accuracy