Computational methods are mathematical models used to study the behavior of complex systems using a computer simulation. Computational models are combining different mathematical and statistical tools that are used to analyze datasets. These computational models help us understand the relationship between a desired outcome and different factors affecting the outcome in order to solve problems. Machine learning is a data analysis method that automates building of analytical models by using algorithms that learn from data and find hidden insights without being programmed where to look. For example, investment banks use computational models to determine how to buy or sell shares in the stock market in order to make a profit or liquidate their holdings in order to avoid losses. Due to dynamic nature of the stock market, investment banks need to make investment decisions quickly, and therefore constantly feed new and updated data to a computer which then uses machine learning algorithms to generate new models. This study analyzes the accuracy of three different computational methods that were generated using machine learning techniques: recursive partitioning, neural networks and deep learning, and Naïve Bayes methods.