Using Commodities to Predict the Swedish Stock Market

This thesis will try to answer the question if it is possible to use commodities to predict the Swedish stock market. The question is answered by searching for an in-sample and out-of-sample predictability relationship between commodity returns and stock returns. Different commodity indices are used in the thesis as predictors in order to predict the general Swedish OMX Stockholm 30 stock index but also to predict eight chosen Scandinavian stocks active in different sectors of the market. This thesis is the first academic paper to do so using an econometrical approach. The thesis starts with an introduction to the stock and commodity market. The link between both markets is discussed and using previous research it is shown that it should be possible to predict stock returns using commodities, at least to some extent. Then follows a theory part where prediction of returns is discussed using the Efficient Market Hypothesis. The hypothesis states that using historical data to predict future returns should be impossible because the market is effective. To test the hypothesis an empirical analysis in several steps follows where it is proved that the hypothesis does not hold since the predictability link between some of the commodity indices used as predictors and the predicted stock returns is strong and robust. The highest level of predictability with a possibility to explain 14.67 % of the one period ahead return in-sample and 13.98 % of the return out-of-sample is achieved using the London Metal Exchange index as a predictor on the Swedish mining company Boliden