Bayesian forecast of monthly forecast in Buenos Aires
Changes in climate variability have affected the frequency of extreme hydrological events. One of the limits of the analysis of these phenomena lies in the missing information, which generates the determination of the analysis methodologies likely to obtain unreliable results. This work presents a Bayesian estimate of the monthly forecast for the city of Buenos Aires collected by the Observatory of the city and published by the National Meteorological Service, the analyzed information corresponding to the monthly data from January 1991 to May 2020.
The function was performed using the rstan programming language, created a connection between the STAN package created in the C ++ programming language and R. The Stan language was used for the statistical model (Bayesian) with an imperative program that computes the log response density function for which we want to make estimates. The prediction of these events that would take place recurring floods in the suburbs of the city of Buenos Aires have generated human and material damages.
Gregorio Saavedra; I am an associate professor of statistics in the Department of Mathematics of the Externado University of Colombia, as well as professional support in the Colombian national navy. My work focuses on innovation in statistics; emphasizing on the programation, reproducible research, open source education and student-centered learning.