Validation of the Prediction of Effectiveness of Statistical Time Series Models using an Artificial Neural Network Model

Authors

Cevallos Torres, Lorenzo

Ochoa Flores, Angel

Chóez Acosta, Luis

Patiño-Pérez, Darwin

Botto-Tobar, Miguel

Icaza Rivera, Dalva

Sarmiento Barreiro, Liliana

Abstract

The objective of this project was to predict the number of cases of infections and deaths from covid-19 through the application of artificial intelligence techniques in order to validate the effectiveness of a statistical model and counteract congestion in the health area within the territory. Ecuadorian, the rapid spread that caused serious consequences in the health systems and the virus triggered a global health crisis, the drastic impact on people's lives caused the application of Artificial Neural Networks-RNA techniques to obtain rapid diagnoses and effective. Historical data from the Ecuadorian state about the infections and deaths recorded per day were taken, the data was processed using the time series statistical method technique and later in the RNA models for the generation of the prediction and validation of the statistical method, the results obtained from each of the neural networks provided a feasible forecast that was close to the real values. The main conclusions show that the techniques applied in this project are efficient when predicting the number of cases of infection and death from covid-19 based on historical data and that the use of neural networks is very useful for solving various predictive problems.