Outbreaks Predictions of Covid19 in Tabasco, Mexico, using Fast Fourier Transform

Abstract

Results of analysis carried out on data of Covid19, in Tabasco State, are presented, from March 2020 to February 2022. The procedure named 7 day´s moving average was applied to smooth the original graph and, in this way, analyze the data more easily. Fast Fourier transform (FFT) was applied to find the frequency that allows us to detect the period (in weeks) of generation of greater Covid19 infections. FFT allows us to determine that for a period of 14n weeks (n=1,3,5,7) there will be a new outbreak. The analysis of data reveals, during the first wave, the week with the highest number of infections was July 7-12 (2751 cases), that is in week 14 ( n=1). FFT indicates that the second wave of infection would be in week 42 with n=3, the maximum was obtained at week 40 (2122 cases). The next prediction (variant Delta) was for week 70 (n=5), where there would be a high number of cases (thrid wave); real data indicates that it was reached in week 73 (7,023 cases). The four-wave (Omicron) is predicted for week 98, however, it was reached in week 94 (12,834 cases). It should be noted that due to the high transmissibility of Omicron, the number of infection grew fast during January 2022, so in this case, the difference between the prediction and the reality, present a difference of four week, however, for week 94 number of cases remains very high (1265 cases). FFT has turned out to be an adequate tool to make predictions of four waves that have occurred in Tabasco.