The purpose of this study is, first, to find out, based on historical data, whether quarterly averages of non-OPEC supply follow a seasonal pattern. If that is mathematically established, then, secondly, it is attempted to estimate the best seasonal factors to decompose the estimated yearly average into seasonal averages.This study applies the Fourier analysis to quarterly supply series to test for seasonality, and provides estimates of seasonal factors for the year 2001 by applying the so-called X-11 decomposition method to the annual estimate. A set of historical data, consisting of quarterly supply averages of individual countries, regional subtotals and aggregate non-OPEC for the period 1971-2000, forms the basis of the analysis.Through the application of the Fourier analysis and X-11 decomposition method, it is demonstrated that quarterly non-OPEC supply, be it by an individual major producer or regional sub-totals, clearly follows a seasonal pattern. This is a very useful conclusion for the market analyst involved with forecasting the quarterly supply.

No abstract

The purpose of this study is to find an efficient method of forecasting average annual non-OPEC supply. Adopting a bottomup approach, it attempts to identify the most suitable forecasting method for each of the world's eight major non-OPEC producers separately. Assuming a short-term forecasting horizon, the study considers supply as an independent variable capable of being forecast fairly accurately, on the basis of historical data. Therefore, the individual countries' average annual production figures for the period 1970-2001 form the basis of the analysis.The eight major non-OPEC producers are classified into three groups, based on the difference between actual production and its five-year moving average during the period . Two different moving average (MA) methods, along with the autoregressive moving average (ARIMA) method, are applied to historical data, to draw forecasts of the average supply for each country for 2000 and 2001. The smoothed absolute percentage error (SAPE) is used as the basis for evaluating (continued overleaf) 126Abstract -continued and comparing the accuracy of various forecasts, and then the most efficient forecasting method is identified.Finally, it is shown that the MA and ARIMA methods are capable of producing remarkably close forecasts for supply by major non-OPEC producers. It is also noted that, in view of the numeracy of factors affecting supply, the best forecasting method for each country does not necessarily remain the same from one year to the next, and that further research, covering a longer period, could end up identifying the single best method.

The objective of this paper is to estimate a weighted variables formula that could be used to calculate OPEC quota distribution. Instead of assigning an arbitrary weight to each of the selected variables included in the formula, we estimate weights based on historical data, using regression analysis. Six national characteristics of OPEC Member Countries, related to oil and socio-economic factors, are considered in the estimation. Time-series are used from 1982 to 2001, for each OPEC Member Country. The results show that the estimated weights are sensitive to the periods considered in the analysis, as well as the number of variables selected. However, there is no specifi c optimum way of dealing with the sensitive, complex quota issue.