This article focuses on the econometric analysis of the prices of oil and gold. The aim is to determine the degree and nature of the investigated commodity dependence in terms of short-term and long-term relationships. The work contains basic characteristics, determinants of price development and theoretical description of statistical tools used to analyze dependencies of investigated time series. In the practical part of the article there is given its own analysis and final interpretation of the development of studied commodities. There are used methods of correlation and regression analysis, Granger causality, Augmented Dickey-Fuller test of stationarity, Johansen test. With respect to Engle-Granger test the two variables have a long run equilibrium relationship. Moreover, the Granger causality test reveals that in long-term, the change in prices of gold influences the change in prices of oil, while the chance in prices of oil does not influence the future change in prices of gold. For time series analysis (monthly average commodity prices, April 1983 - December 2016) there was used computer program GRETL.