Stock market change prediction based on multiple linear regression and CNN-LSTM

Abstract

The impact of climate change on the economy and society has become increasingly evident with the arrival of the 21st century. This impact has extended beyond the realm of economic development and has also affected the financial sector. In consideration of this, the article develops a set of models following a thorough analysis of data released by the National Bureau of Statistics of China, in order to investigate the relationship between climate change and stock indices. The article employs the Spearman correlation coefficient to examine the intricate connection between climate change and stock indices. This article aims to comprehensively analyze the potential impact of climate change on stock index trajectories through the establishment of multiple linear regression, ridge regression, and LASSO regression models. The objective is to facilitate the formulation of well-founded predictions. Finally, the article anticipates the future prices of oil and gas stocks for the next 100 days through the implementation of a CNN-LSTM model. The integration of Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) architecture is anticipated to enhance the precision and adaptability of stock price trend forecasts within the context of dynamically evolving environmental conditions.