Modelling Commodity Prices in a World Macroeconomic Model
This paper describes work in progress to endogenise commodity prices in the OECD’s world macro—model INTERLINK. Part I gives the purpose of the work and an overview of the model and describes the modelling strategy. Part II gives results of empirical work on commodity price equations. Long run properties of estimated equations are reported, together with short run forecast performance. Simulation testing of the full model system incorporating the equations is still in progress at the time of writing and only preliminary results are available.
KeywordsCommodity Price ARIMA Model Import Price Spot Price Real Price
Unable to display preview. Download preview PDF.
- 1.See G.H. Holtham (1984), ‘Multinational Modelling of Financial Linkages and Exchange Rates’, OECD Economic Studies, 2, for a description of this part of the system. It has subsequently been revised but the approach remains similar.Google Scholar
- 3.Beckerman, W. and T. Jenkinson, ‘Commodity Prices, Import Prices and the Inflation Slowdown: A Pooled Cross-country Time Series Analysis’, unpublished mimeo.Google Scholar
- 4.A previous attempt to endogenise commodity prices in the context of a world model was the COMLINK project. A series of 28 commodity market models ranging from two to over 30 equations were integrated into the Project Link system. This was a more elaborate and ambitious approach than the present one. See F.G. Adams, ‘Integrating Commodity Markets into LINK’ in LA. Sawyer, Modelling the International Transmission Mechanism, North Holland, New York.Google Scholar
- 5.This work, by Mr. Paul Saunders of OECD, is reported in a forthcoming OECD Economics Department Working Paper, ‘Commodity Prices in INTERLINK’.Google Scholar
- 6.See Monthly Commodity Price Bulletin, UNCTAD, Geneva and Supplement TD/B/C.1/CPB/L.24/ADD.1, April 1983. These monthly prices are calculated from daily or weekly quotations. They have been averaged to give semestrial series.Google Scholar
- 7.See H.J. Timm (1980), ‘A New HWWA Raw Material Price Index’, Hamburg. Monthly prices are based on daily computations and they are averaged to give semestrial series. The HWWA has kindly agreed to supply the OECD with an unpublished index weighted according to OECD exports, and this may be used in future developments.Google Scholar
- 8.See Wickens, M.R. (1982), ‘The Efficient Estimation of Econometric Models with Rational Expectations’, Review of Economic Studies.Google Scholar
- 9.See D. Ridler and C.A. Yandle (19), ‘A Simplified Method for Analysing the Effects of Exchange Rate Changes on Exports of Primary Commodity’, IMF Staff Papers, 19.Google Scholar
- 10.Although the frost was in mid—1975 its full effects on supply were not realised until they affected Brazilian production in the 1976–77 season — roughly March to March. For FOOD the estimated value of the dummy was 0.3 for 1973I and II, 1 for 1974I and II and 0.4 for 1980I and II, 0 otherwise; for TBEV it was 1 for 1976I to 1977I, 0 otherwise.Google Scholar
- 13.Alternative regression basd techniques were first proposed by Granger, C.W.J. (1969), ‘Investigating Causal Relationships by Econometric Methods and Cross Spectral Methods’, Econometrica, 36, and Sims, C. (1972), ‘Money, Income and Causality’, American Economic Review, 62.Google Scholar
- 14.Sims, C. (1977) in ‘Comment on Pierce’ Journal of the American Statistical Association, 72, No. 357, pp. 23–24.Google Scholar
- 15.Chu, Ke-Young and T.K. Morrison (1984), ‘The 1981–82 Recession and Non-oil Primary Commodity Prices’, IMF Staff Papers (March). Grilli, E.R. and Maw-Cheng Yang (1981), ‘Real and Moneatry Determinants of Non-oil Price Movements’, Division Working Paper No. 1981–6, Commodities and Export Projections Division, The World Bank.Google Scholar