Does cultural economics affect country’s competitiveness?

Abstract

Stock market prediction is a challenging task due to the nature of financial markets which are volatile and non-linear. Traditional machine learning models often struggle to capture the complex temporal dependencies and underlying patterns in stock price movements. This paper presents a hybrid model that combines Long Short-Term Memory (LSTM) networks with eXtreme Gradient Boosting (XGBoost) to enhance stock market prediction accuracy. LSTM, a type of recurrent neural network (RNN), is well-suited for handling time series data, as it effectively models sequential dependencies. Financial time-series data are known to be volatile, unpredictable, non-linear and temporally dependent making stock market prediction a tough problem. A hybrid model of Long Short-Term Memory (LSTM)-XGBoost is proposed to improve the stock closing price prediction by using the sequential learning ability of LSTM and the non-linear modelling ability of XGBoost. Historical stock-market data for Alibaba (BABA) from 1 January 2016 to 1 October 2021 was used for the analysis. The data collection comprises important market characteristics such as Open, High, Low, Close, Adjusted Close, and Volume. The LSTM model learned the temporal relationships and patterns from the historical stock-price sequences. Then the residual errors and other non-linear relationships were modelled to improve the LSTM predictions with the XGBoost. Model Evaluation The model performance was evaluated using Mean Squared Error (MSE), Mean Absolute Error (MAE) and R2. The findings reveal that the standalone LSTM model achieved an MSE of 0.1219, MAE of 0.3163, and R2 of −4.1150. The hybrid LSTM-XGBoost model achieved an MSE of 0.0001, MAE of 0.0050, and R2 of 0.9972. The results show that the prediction metrics improved significantly by adding XGBoost to the LSTM framework. The study shows the potential of hybrid deep learning and boosting techniques in stock market prediction.

Keyword : RNN, LSTM, XGBoost, stock market, prediction, neural networks, accuracy, precision