StockBot 2.0: Vanilla LSTMs Outperform Transformer-based Forecasting for Stock Prices

TL;DR

This study demonstrates that vanilla LSTM models outperform transformer-based models in stock price forecasting, emphasizing the robustness and data efficiency of recurrent neural networks in financial time-series prediction.

Contribution

The paper introduces an enhanced StockBot architecture and provides comprehensive empirical evaluation showing LSTMs outperform transformers in stock forecasting tasks.

Findings

Vanilla LSTMs achieve higher accuracy than transformer models.

LSTMs provide more stable buy/sell signals.

Recurrent models are more data-efficient in this context.

Abstract

Accurate forecasting of financial markets remains a long-standing challenge due to complex temporal and often latent dependencies, non-linear dynamics, and high volatility. Building on our earlier recurrent neural network framework, we present an enhanced StockBot architecture that systematically evaluates modern attention-based, convolutional, and recurrent time-series forecasting models within a unified experimental setting. While attention-based and transformer-inspired models offer increased modeling flexibility, extensive empirical evaluation reveals that a carefully constructed vanilla LSTM consistently achieves superior predictive accuracy and more stable buy/sell decision-making when trained under a common set of default hyperparameters. These results highlight the robustness and data efficiency of recurrent sequence models for financial time-series forecasting, particularly in the absence of extensive hyperparameter tuning or the availability of sufficient data when discretized to single-day intervals. Additionally, these results underscore the importance of architectural inductive bias in data-limited market prediction tasks.