Forecast of selected S&P 500 stock prices

This thesis is focused on the research of the weak form of the efficient market hypothesis for randomly selected stocks, which belong under the stock index S&P 500. The content of this work is the investigation of the possibility that the stocks, which are being part of the stock index S&P 500, are struggling with the weak market inefficiency, which could for example be caused by the fact that this index is being widely used by not only the classical large investors, but also the small retail ones which tend to have less experience with investing. For this purpose, there are being used two model types. The first on is the linear stochastic model ARIMA. The second type of models are neural networks, more specifically Long Short Term Memory models also known as LSTM. These two model types are being compared, especially in their predictive abilities for examined data. For the purpose of examining and comparing the results of particular model types, there will be proposed a simple strategy, with whose help it will be possible to compare and evaluate the predictive abilities of these models. My main focus on the theoretical part is the description of the used model types and the information cennected with the efficient market hypothesis and the stocks index S&P 500.