Implementation of Statistical Tool in Legendary Pokémon Analysis and the Pricing in Stock Market

Abstract

Statistics are widely applied in wide spread of advanced science researches to give the quantitative crustily, which is now much more well-liked. In this paper, a dataset is analyzed and a pricing method is presented based on two different statistical models. In the random forest model, we first do a simple statistical data analysis to determine the association between attributes and "is legendary" using the Legendary Pokémon dataset. The above analysis results may then be used to determine which characteristics can make a Pokémon legendary. In the Black-Scholes model, we first discuss SDE before using it to provide a stock pricing method. For the former one, the statistical data analysis in the first application may determine that all the characteristics have a positive correlation with "is legendary," and lastly, random forest discovered that the characteristics that can turn a Pokémon legendary are special attack and speed. Through the second application, it can determine how an SDE’s trajectory and coefficients are related. The pricing method is then obtained by applying SDE to the Black-Scholes model. We value a company’s shares using this method and contrast the accuracy of this approach with more conventional approaches. According to data analysis, the first application easily and intuitively demonstrates random forest’s practical application capabilities, as well as its ease and accuracy for us. These results shed light on guiding further exploration of the practice of random forests in data analysis and the application of SDE related models.