Advanced Strategies for Enhancing Tesla Stock Price Prediction

Pawar, Swapnil Sanjeev (2025) Advanced Strategies for Enhancing Tesla Stock Price Prediction. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (851kB) | Preview
Preview	PDF (Configuration Manual) Download (538kB) | Preview

Abstract

In this paper, a deep learning hybrid model was introduced utilizing both historical market data and sentiment signals collected from social media to predict Tesla stock prices. The architecture involved a combination of Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (Bi-LSTM), and Transformer encoders for extracting local, temporal, and global contextual features from the multivariate timeseries data. VADER sentiment analyser was employed to extract sentiment features from Kaggle datasets of Tesla posts from Twitter and Reddit before processing. The sentiment scores were then accumulated on a daily basis, and combined with technical indicators (such as Open, High, Low, Close and Volume) to form 15-day lookback sequences to predict next day stock prices. The models were trained with MSE and scored with MAE. Individual CNN-BiLSTM and Transformer models were developed, producing MSEs of 0.00318 and 0.02180 while MAEs were 0.04498 and 0.12434 respectively. Based on the insights gained from these results, a final hybrid model was formulated and optimized using Bayesian hyperparameter tuning. The combined model achieved the lowest MAE of 0.04486, MSE of 0.00324 and validation loss of 0.00323 outperforming both baseline models. Sentiment data were integrated through an attention mechanism, which was observed to enhance predictive accuracy. SHAP and LIME were used to interpret the model’s predictions, producing explainable, sentiment aware forecasts for data-driven decision-making.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Sahni, Anu UNSPECIFIED
Uncontrolled Keywords:	CNN; Bi-LSTM; MAE; MSE; SHAP; LIME; VADER; Transformer; Bayesian Optimization
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Motor Industry H Social Sciences > HG Finance > Investment > Stock Exchange
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	18 Nov 2025 17:34
Last Modified:	18 Nov 2025 17:41
URI:	https://norma.ncirl.ie/id/eprint/8943

Actions (login required)

View Item