Copper econometric models currently existing in the open literature incorporate the demand side of the market only as an aggregated demand function. In this thesis, copper demand was disaggregated into its end-use industrial sectors and linear demand equations estimated for each sector. The correlation among the error terms of the various sectoral equations was explicitly taken into account in the estimation. Elasticities were computed at the means for the price and activity variables. A comparison with results using the Ordinary Least Squares method is also provided. Exogenous variables used in each sectoral equation were forecast separately and copper demand was subsequently forecast for each end-use sector. / Master of Arts