A Neutrosophic Statistic Method to Predict Tax Time Series in Ecuador

Prediction of tax collection behavior is an essential tool for social planning by the State of any country. The tax is the State's mechanism for budget collection, which is necessary to accomplish public services that benefit the whole society. This paper firstly aims to propose a method of predicting time series where values can be given in form of intervals rather than numbers. This form permits to obtain more truthful results, but with a greater indeterminacy. Because statistical prediction methods are used, where data in form of intervals are included, we can classify this approach as a kind of Neutrosophic Statistics technique. Basically, the method converts a set of predicted numerical values into intervals. The second objective is to apply the method to predict the monthly income from taxes in Ecuador for the year 2019. Keywords: time series, tax, prediction, neutrosophic statistic, autocorrelation, median average