Open AccessMexican Stock Exchange Performance after the Crisis of 2008: Application of Data Mining
Author(s) -
Eliseo Ramírez Reyes,
Arturo Morales Castro,
Néstor Juan Sanabria Landazábal
Publication year - 2020
Publication title -
dimensión empresarial/dimensión empresarial
Language(s) - English
Resource type - Journals
eISSN - 2322-956X
pISSN - 1692-8563
DOI - 10.15665/dem.v18i(1).2246
Subject(s) - econometrics , artificial neural network , stock exchange , mean absolute percentage error , exchange rate , stock (firearms) , linear regression , index (typography) , statistics , stock market index , computer science , economics , mathematics , artificial intelligence , engineering , geography , stock market , monetary economics , finance , mechanical engineering , context (archaeology) , archaeology , world wide web
Different prediction models are explored to analyze the performance of the Mexican Stock Exchange (PQI) after the 2008 crisis. These models have demonstrated a good prognostic capacity for both multivariable and univariable approaches given their non-parametric characteristics. The selected variables were: Dow Jones Industrial Average Index (DJIA), CPI, International Reserves (IR), CETES28, USDMX exchange rate, (M1) and the sovereign default risk of Mexico (MRDS). The models were evaluated with MAPE and compared with linear regression models (LR) and neural networks (NN). The results show that the models have a similar performance according to the percentages of error they presented.