Predictive analysis for calculating the valuation of the affiliated fund of a private pension system using machine learning techniques and tools
![Thumbnail](/bitstream/handle/10.18687/20190101_343/FP343.pdf.jpg?sequence=6&isAllowed=y)
View/ Open
Date
2019-07Author
Armas, Jimmy
Espinoza Ladera, Jhonatan
Dueñas Castillo, Brian
Aguirre Mayorga, Santiago
Metadata
Show full item recordAbstract
This paper proposes a model for the analysis of the prediction of the accumulated affiliated fund based on an area of study of machine learning. The model allows to project the pension fund of an affiliate to the private pension system by means of a web solution, in order that people have information and adequate tools that allow them to have a general vision of the valuation of their funds over the years until the time of retirement. In Peru, the decree of law 1990 indicates that the year of retirement is 65 years, although there is also the figure of early retirement. The proposed model consists of the use of data analytics based on the modeling of machine learning algorithms through cloud platforms. The structure of the model includes four layers: the transformation of the affiliate's data, the security and privacy of the personal data, obtaining and management of data, and finally, the life cycle of the data applied to the analytics. The model emphasizes data analytics concepts where large amounts of data are examined that lead to conclusions for better decision making. For this, the machine learning technique "boosted decision tree" is used due to the proximity of this technique applied in the financial projections. The model was validated with a pension fund administrator (AFP) in Lima (Peru) and the results obtained focused on the use of improved decision tree regression with a coefficient of determination of 99.997% and an average square error of 0.00650%. The coefficient of determination is an indicator of the quality of the model to predict results while the quadratic error quantifies the percentage of error among the set of results obtained under the boosted decision tree regression model.