ARIMA model on the consumption of water for population use in the city of Tacna.

Abstract

The objective was to build a time series forecast model based on endogenous patterns and variables of drinking water consumption and additionally determine the trend, seasonality, cyclical patterns and characteristics of the water consumed in the city of Tacna for the obtaining forecasts.

The research was non-experimental, correlational and longitudinal, with monthly information from January 2006 to March 2018 recorded through documentary analysis. The unit of analysis were the economic units of the city of Tacna with drinking water service connected to the public network and monthly consumption as variable of interest. The population corresponds to a finite series of monthly data of size N = 383 months. The sample consisted of 139 observations between January 2006-July 2017 with which the ARIMA model has been built based on the Box-Jenkins methodology and extended until March 2018, for the validation of the model.

The main conclusion is the obtaining of the forecast model for the consumption of water for population use in the city of Tacna ARIMA (0,1,1) (0,1,1) 12 with intercept and whose empirical mathematical formulation is:

(LnX) ̂_t=0,000415+0,590a_(t-1)+0,899a_(t-12)

with trend and seasonality patterns.