Violence in Ecuador: Analysis of Homicides Through Time Series

Violence in Ecuador is unprecedented, the rates of homicides, femicides, robberies, attacks and other types of crimes have increased alarmingly in this country. Every day new events are reported that are news and alarm the community. Government institutions, together with the police organization and the military forces, carry out actions to mitigate the wave of violence without achieving efficient results. This paper analyzes the number of homicides nationwide as a time series to know the behavior of this variable from January 2014 to May 2022. Applying smoothing models, as well as the ARIMA and Neural Network models, the most efficient model is sought. that minimizes the prediction error. Among the smoothing models, the Holt model was found to be the most efficient, however, but when comparing all the applied models, it was found that the Neural Network model is the most efficient with which good forecasts can be obtained.

• 第一著者: Vivas Paspuel, Atal Kumar (author)
• その他の著者: Vivas Paspuel, David Alfredo (author)
• フォーマット: article
• 言語: spa
• 出版事項: 2024
• 主題: pronósticos; series de tiempo; modelos; homicidios; forecasts; time series; models; homicides
• オンライン･アクセス: https://revistadigital.uce.edu.ec/index.php/INGENIO/article/view/4546