a factor model for world trade growth
play

A FACTOR MODEL FOR WORLD TRADE GROWTH Elena Rusticelli and Stphanie - PowerPoint PPT Presentation

A FACTOR MODEL FOR WORLD TRADE GROWTH Elena Rusticelli and Stphanie Guichard OECD Economics Department 6 th Colloquium on Modern Tools for Business Cycle Analysis Luxembourg 26-29 September 2010 Outline of the presentation 1. World trade


  1. A FACTOR MODEL FOR WORLD TRADE GROWTH Elena Rusticelli and Stéphanie Guichard OECD Economics Department 6 th Colloquium on Modern Tools for Business Cycle Analysis Luxembourg 26-29 September 2010

  2. Outline of the presentation 1. World trade growth forecasting: overview and motivation of the study 2. Key indicators 3. Short-term forecasting methods 4. Methods comparison and empirical results 5. Concluding remarks

  3. Two different approaches to forecast world trade � The world trade growth is traditionally forecasted using a bottom-up approach where import and export volumes are forecasted on a country basis and the forecast for world trade is simply the aggregation of country-specific forecasts (OECD, IMF, WTO, World Bank). � Short-term forecasting methods, corresponding to the direct approach, are normally considered as a useful benchmark which can point to possible up-downside risks to the current projections. � Some studies (e.g. Burgert and Dées 2008) show the superior performance of direct forecasting methods where global factors play a fundamental role to explain world trade beside the traditional country-specific determinants.

  4. The OECD uses the bridge model as a complement to the bottom approach Trillions of 2005 US dollars 16 15 14 13 12 2006 2007 2008 2009 2010 Current world trade series EO87 projections (bottom-up approach) Bridge models projection (direct approach)

  5. Motivation of the study � The main OECD tool for short-term forecasting of world trade growth - the bridge equation model - has been complemented with a new dynamic factor model which allows to extend the dataset - to include a larger number of monthly series on world and country level or different levels of aggregation without the risk of multicollinearity , losses of degrees of freedom and the increase in the computational burden - to include relevant indicators for explaining world trade which are available only with a shorter history - to evaluate the contribution that different indicators have in the final forecasts, as well as their lagging or leading properties.

  6. The key variables to forecast world trade � A large unbalanced dataset of 35 monthly indicators of different nature - hard, soft and financial indicators. � Different levels of aggregation – global and country level, aggregate or disaggregate components. � Stationarity achieved by means of monthly growth rates for all hard indicators ( except the Baltic Dry index) and the world share prices. Among survey indicators only the world stock level index has been transformed with first order differences.

  7. Starting Publication Monthly Indicators Source date lags HARD INDICATORS ECONOMIC ACTIVITY World industrial production index 1991 2 CPB USA industrial production index 1991 2 CPB Japan industrial production index 1991 2 CPB Euro area industrial production index 1991 2 CPB Advanced economies industrial production index 1991 2 CPB Emerging economies industrial production index 1991 2 CPB 2 Asia industrial production index 1991 CPB Latin Ameria industrial production index 1991 2 CPB Central and Eastern Europe industrial production index 1991 2 CPB Africa and Middle East industrial production index 1991 2 CPB Largest countries industrial production index 1990 2 OECD calculations OECD OECD retail sales 2000 3 IISI World steel production 1980 1 SHIPPING AND FREIGHT ACTIVITY The Baltic Exchange Baltic dry index 1985 1 Harpex shipping index 1996 1 Harper Petersen & Co. International air traffic 1996 2 IATA GLOBAL TECHNOLOGY CYCLE Tech pulse index 1971 1 CSIP SIA World semiconductor billings 1976 2 TRANSPORT COSTS Brent oil prices 1957 1 UK Dept. of Energy SOFT INDICATORS EXPORT ORDERS G7 export orders 1962 1 OECD calculations ISM World export orders 1998 1 PURCHASING MANAGERS'INDEX Global PMI index 1998 1 ISM ISM PMI stock level index 1998 1 OECD + BRICS CLI 1960 2 OECD FINANCIAL INDICATORS Datastream World stock market prices index 1973 1 US high yield spread 1984 1 OECD calculations FED US loan officer survey (quarterly) 1990 1

  8. Ranking of indicators – including contemporaneous Whole sample Sample ending in 2008 Q2 SIC SIC Adj. R 2 Adj. R 2 max lag value max lag value Ranking Ranking World industrial 0 0.91 -7.35 1 0 0.80 -7.60 1 production (CPB) World export orders 4 0.91 -6.61 2 2 0.69 -6.67 3 Largest countries 0 0.86 -6.94 3 0 0.62 -6.98 6 industrial production Global PMI index 2 0.84 -6.22 4 2 0.69 -6.68 2 Air freight volume 0 0.80 -6.26 5 0 0.66 -6.83 5 OECD+BRICS CLI 1 0.79 -6.51 6 1 0.67 -7.07 4 G7 export orders 2 0.75 -6.27 7 1 0.49 -6.64 8 US high yield spread 4 0.74 -6.14 8 0 0.39 -6.50 14 World stock market price 1 0.69 -6.11 9 -1 0.49 -6.63 9 Baltic Dry Index 2 0.65 -5.96 10 0 0.33 -6.41 18 OECD retail sales 2 0.65 -5.94 11 2 0.44 -6.50 11 World steel production 1 0.61 -5.87 12 0 0.34 -6.42 16 Semi computers billings 0 0.57 -5.83 13 0 0.46 -6.62 10 PMI stock level index 0 0.54 -5.30 14 1 0.51 -6.29 7 US loan officer survey 2 0.53 -5.65 15 0 0.41 -6.53 13 US tech pulse index 0 0.52 -5.72 16 0 0.43 -6.57 12 Oil prices 0 0.52 -5.71 17 0 0.37 -6.47 15 Harpex index 0 0.42 -5.52 18 0 0.34 -6.41 17 Max lag is based on the Schwarz criterion value, but ranking were not affected by changing the lag selection criteria

  9. Ranking of indicators – excluding contemporaneous Whole sample Sample ending in 2008 Q2 SIC SIC Adj. R 2 Adj. R 2 max lag value Ranking max lag value Ranking OECD+BRICS CLI 4 0.81 -6.47 1 2 0.58 -6.76 1 World stock market price 1 0.61 -5.91 2 1 0.49 -6.67 2 Baltic Dry Index 2 0.61 -5.87 3 1 0.33 -6.41 17 US high yield spread 3 0.60 -5.81 4 2 0.41 -6.48 7 Global PMI index 2 0.59 -5.33 5 2 0.46 -6.18 3 World export orders 2 0.58 -5.31 6 2 0.44 -6.15 5 Air freight volume 1 0.57 -5.48 7 1 0.41 -6.27 6 World industrial production (CPB) 1 0.56 -5.76 8 1 0.38 -6.45 10 OECD retail sales 2 0.55 -5.73 9 2 0.39 -6.45 9 G7 export orders 2 0.54 -5.70 10 2 0.40 -6.47 8 US loan officer survey 2 0.49 -5.59 11 1 0.37 -6.48 11 PMI stock level index 1 0.46 -5.03 12 1 0.44 -6.22 4 Largest countries industrial production 2 0.46 -5.54 13 1 0.35 -6.43 14 World steel production 1 0.45 -5.58 14 1 0.33 -6.41 18 Oil prices 2 0.44 -5.50 15 1 0.33 -6.41 16 US tech pulse index 2 0.42 -5.47 16 1 0.35 -6.44 13 Harpex index 1 0.36 -5.42 17 1 0.33 -6.41 15 Semi computers billings 1 0.35 -5.42 18 1 0.37 -6.46 12 Max lag is based on the Schwarz criterion value, but ranking were not affected by changing the lag selection criteria

  10. Best coincident indicators World industrial production index World exports orders .06 .04 60 .04 .04 .02 56 .02 .02 .00 52 .00 .00 -.02 48 -.02 -.02 -.04 44 -.04 -.04 -.06 40 -.06 -.06 -.08 36 -.08 -.08 -.10 32 -.10 90 92 94 96 98 00 02 04 06 08 90 92 94 96 98 00 02 04 06 08 Largest countries industrial production index Global PMI index 64 .04 .06 .04 60 .02 .04 .02 56 .00 .02 .00 52 -.02 .00 -.02 48 -.04 -.02 -.04 44 -.06 -.04 -.06 40 -.08 -.06 -.08 36 -.10 -.08 -.10 90 92 94 96 98 00 02 04 06 08 90 92 94 96 98 00 02 04 06 08

  11. Best leading indicators OECD+BRICS CLI World stock market price index 104 .04 .3 .04 .2 .02 103 .02 .1 .00 102 .00 .0 -.02 101 -.02 -.1 -.04 100 -.04 -.2 -.06 99 -.06 -.3 -.08 98 -.08 -.4 -.10 97 -.10 90 92 94 96 98 00 02 04 06 08 90 92 94 96 98 00 02 04 06 08 Baltic Dry index US high yield spread 14,000 .04 18 .06 16 .04 12,000 .02 14 .02 10,000 .00 12 .00 8,000 -.02 10 -.02 6,000 -.04 8 -.04 4,000 -.06 6 -.06 2,000 -.08 4 -.08 0 -.10 2 -.10 90 92 94 96 98 00 02 04 06 08 90 92 94 96 98 00 02 04 06 08

  12. Short-term forecasting methods applied on macroeconomic aggregates � Random walks and autoregressive models � Quarterly VARs (Sédillot and Pain 2003, 2005) � Bridge equations models (Sédillot and Pain 2003; Baffigi et al. 2004) � Diffusion indices (Stock and Watson 2002) � Dynamic factor models (Forni et al. 2007; Ba ń bura and Rünstler 2007)

  13. Overview of the OECD bridge equation model � Monthly indicators dataset includes : - world IP index - G7 countries export orders - the two technology indicators (semiconductor billings and tech pulse index) - oil prices - Baltic dry index � Quarterly indicators dataset includes: - US loan officer survey - OECD world trade growth of goods and services � Four quarterly forecasts for world trade growth produced: backcast, nowcast and one-two quarters ahead forecasts.

Recommend


More recommend