Predicting stock return and volatility with machine learning and econometric models - a comparative case study of the Baltic stock market

For stock market predictions, a crucial problem is predicting the prices as accurately as possible. There are different approaches (for example, econometrics and machine learning) for predicting stock returns. However, it is non-trivial to find an approach which works the best. In Baltic countries, the interest in stock market investment has grown due to the popularisation of investment and the favourable trading conditions of banks, however, there are very few attempts to apply machine learning and neural network models in the studies of the Baltic region. In this paper, we make a thorough analysis of the predictive accuracy of several machine learning and econometric approaches (ARMA, GARCH, random forest, SVR, KNN and GARCH-ANN) for predicting the returns and volatilities on the OMX Baltic Benchmark Price Index. Our results show that the machine learning methods predict the returns better than autoregressive moving average models for most of the metrics.