Predictive Analytics of Tesla and Google Stock Prices Using Advanced Machine Learning Frameworks

View at publisher Summarise with VIRAT
Authors: Ganeshvel K P, Harshavardhana D, Ponnaratta Gopika
Cited by: 0
References: 16

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Stock market forecasting remains a challenging task due to the nonlinear, volatile, and stochastic nature of financial time- series data. This study aims to (i) comparatively evaluate statistical and machine learning approaches for stock price prediction, (ii) assess the impact of technical feature engineering on forecasting accuracy, and (iii) analyze the suitability of different models for volatile technology-driven stocks. A comparative framework is developed using ARIMA, Prophet, and Random Forest regression to forecast the daily closing prices of Tesla and Google stocks. Historical price data are preprocessed and enriched with technical indicators, including moving averages, momentum indicators, and volatility measures. Model performance is evaluated using Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Coefficient of Determination (R2). Experimental results demonstrate that the Random Forest model consistently outperforms ARIMA and Prophet, achieving lower prediction errors due to its ability to capture nonlinear relationships through engineered features. ARIMA and Prophet serve as effective statistical baselines for trend and seasonal analysis. The findings confirm that feature-driven ensemble learning significantly enhances predictive accuracy, highlighting its applicability in algorithmic trading and decision-support systems.

Fields of Study & Subjects

Stock Market Forecasting Methods Forecasting Techniques and Applications Financial Markets and Investment Strategies

Publication details

Journal / source
International Conference on Intelligent Sustainable Systems (ICISS)
Pages
1849–1856
Publication date
4 March 2026
Language
EN