Predicting Twitter stock price using linear regression, random forest, and MLP regression

Abstract

Since the chaos and complexity in the stock market, predicting stock prices with machinery approaches helps assert a new perspective for people to reach satisfactory results in stock analysis. The stock price of Twitter reflects its market value and performance, which are influenced by various factors such as user engagement, revenue, news, and sentiment. Predicting the stock price of Twitter is a challenging task that requires sophisticated methods and data analysis. In this essay, the author compares three different machine learning models to predict Twitters stock price: linear regression, random forest, and MLP regression. The paper uses historical data on Twitters stock price and various features such as volume, peak value, and trough value. The researcher evaluates the performance of each model using metrics such as Mean-squared Error and R-squared and finds that MLP regression outperforms the accuracy and generalization of the other two models. The author also discusses the limitations and implications of our findings for investors and researchers.