Analysis of Macroeconomic Factors in Stock Market

This paper is a continuation of a study initiated in the course work and dedicated to predicting the Micex Index using neural networks. The main goal of this paper is to analyze macroeconomic and other factors affecting the stock market dynamics with the following application of the results to build econometric models of the GARCH family and LSTM neural networks that can predict daily values of the Micex Index. As part of the study, it was revealed that Gross savings, Unemployment, Gross Domestic Product, Exports and Incomes Per Capita have the greatest impact on the stock market. Among other factors, the S&P-500 index, the dollar rate, and the price of Brent crude have the most influence. Models of the GARCH family are better able to predict based on influencing factors. Neural networks demonstrate the best predictive power on historical quotes for the past 30 trading days and are ahead of the GARCH family econometric models in forecasting quality. The results of this study will be useful for conducting a fundamental analysis of the stock market, and for the practical development of trading strategies.