Code and program for analyzing STI Data for the analysis of Sexually Transmitted Infections.

Through ARIMA modeling and dynamic visualizations, this project reveals how external factors like weather and holidays shape bike rental demand. The insights support smarter staffing and inventory planning, with a focus on making data-driven decisions accessible to both analysts and business teams.

Este proyecto explora cómo la cantidad de capas y neuronas en una Red Neuronal afecta la precisión del pronóstico de generación solar, eólica y demanda eléctrica. Se implementaron modelos en R con la librería neuralnet, evaluando arquitecturas como A1 = c(3), A2 = c(3,3), A3 = c(3,3,3), A4 = c(3,3,3,3), A5 = c(3,3,3,3,3), A6 = c(3,3,3,3,3,3), A7 = c(6), A8 = c(8), A9 = c(10), A10 = c(12), A11 = c(15,15), A12 = c(10,10), A13 = c(20). Se compararon las métricas de error (R², RMSE, MAE) para cada arquitectura, observando el impacto de aumentar la profundidad y el número de neuronas sobre la precisión del pronóstico. Los resultados muestran el desafío de equilibrar la complejidad del modelo con la precisión en series temporales de energía.

In this report we aim to describe the effects of storms across the United States on population health and the economy, between January 1950 to November 2011. Our overall hypothesis is that certain storm event types have a noticeably more negatitve impact on either population health or the economy. To investigate our hypothesis, we obtained USA storm data orginally from the US National Oceanic and Atmospheric Administration (NOAA), downloaded from the [Cousera course website](https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2) for the period January 1950 to November 2011.