Moc-Teml: caso de estudio predicción de tendencia en los índices bursátiles de la Bolsa de Valores de Colombia

Abstract

This project investigates supervised learning techniques from a branch of artificial intelligence (AI) known as Machine Learning, to put into practice the application of two algorithms, Support Vector Machines (SVM) and artificial neural networks (ANN), which allow generating a comparison of precision results in the predictions of success on the path followed by the prices of the stock market indices of the Colombian Stock Exchange (BVC), what is commonly known as a trend. As a result, a document is delivered with the results of the comparative analysis of the models applied, where it is possible to identify the levels of deviation and proximity in the certainty of the results of each model. The representative advantages of each model vs. the others applied are also mentioned. So that it is possible to guide whoever is interested, on which of the two algorithms has the highest degree of certainty.