THE GENERAL SCHEME FOR DETERMINING THE PARAMETERS OF THE ARIMA MODEL IN ECONOMIC FORECASTING

Dmitri TERZI Universitatea de Stat din Moldova

Abstract

In this paper we consider the time series that are determined under the conditions of invariance of the mutual distribution of the probability of observation as a condition for building the ARIMA model. The following are presented: a) general scheme for determining the parameters of the ARIMA model based on covariance between different time series with different lags; b) time series modeling programs for models testing; forecasting recommendations based on information from various examples. Keywords: ARIMA, time series, stationarity, forecast, covariance, autocorrelation and partial autocorrelation, general scheme, parameter determination and model testing.