Villacencio Mattos, John

Profile Picture

Filters

20101
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Publication
    Ciclos económicos de Puerto Rico: uso de modelos y técnicas estadísticas multivariadas
    (2010-05) Villacencio Mattos, John; Quintana Díaz, Julio C.; College of Arts and Sciences - Sciences; Lorenzo González, Edgardo; Alameda, José; Department of Mathematics; Hunt, Shawn
    Economic cycles provide the best monthly economic indicators of a country, both on the long and short term scale. For this reason it is necessary that the cycles model the economy to the most optimal degree possible. This research project proposes a new model for the economic cycles in Puerto Rico. This model, based on the findings of our study, works better than the previous system because it eliminates the problems of multicollinearity and autocorrelation. This study does a comparative analysis of different techniques such as Discriminant Analysis and Logistic Regression to examine which of these methods better predicts economic cycles. According to the Discriminant Analysis method the variables “Motor Vehicle” and “Salary of the Manufacturer” are those that contribute significantly to discrimination of other groups, while with Logistic Regression we found that the variables “Motor Vehicle”, “Sale of Cement” and “Salary of the Manufacturer” were those that contributed to the discrimination. In conclusion, this study would seem to indicate that the Discriminant Analysis method is preferable to the other because of the fact that this technique results in a lower rate of poor classification.