Accurate stock price prediction remains a challenging task due to the inherent complexities and high volatility of financial markets. Traditional models like ARIMA often fail to capture non-linear dynamics, while standalone deep learning models such as GRU and LSTM, though effective, may not fully leverage the benefits of combining technical and fundamental analysis. This paper proposes an integrated GRU-LSTM model that incorporates both technical indicators (e.g., MACD, RSI) and fundamental data (e.g., P/E ratio, profitability) to enhance predictive accuracy. The model is evaluated on two Chinese stocks, 600,719.SS and 000679.SZ, using historical data from Yahoo Finance. Experimental results demonstrate that the integrated GRU-LSTM model outperforms traditional ARIMA and standalone GRU/LSTM models, achieving RMSE values of 0.0141 and 0.0360 for the respective stocks. The study also explores the impact of varying GRU and LSTM layer configurations, finding that balanced architectures yield the best performance. This research highlights the potential of combining technical and fundamental analysis within a deep learning framework for more accurate stock price prediction.