Exchange Rate Uncertainty and Trade Growth - A Comparison of Linear and Nonlinear (Forecasting) Models



Our analysis focuses on the one step forecasting horizon. As direction of future research it
appears fruitful to also compare alternative approaches with respect to forecasting trade
growth conditional on FX uncertainty at longer horizons. In this case, considering medium
and long term measures of volatility becomes particularly imp ortant.

A DATA

A.1 Computation of data series

The computation of real effective FX rates or foreign economic activity requires an appropri-
ate weighting scheme taking the relative importance of trading partners for each member of
the cross section into account. Most appropriately, one would refer to sectoral trade flows to
quantify the sector specific importance of a country as trading partner. For the wide range
of countries analyzed in this study bilateral sectoral trade data is unavailable. To this end,
we follow the procedure in Klein
16 and base the computation of weights on aggregate trade
data that are available on a bilateral basis. This implies that weights are constant across
sectors. The weight w
kl attached to a partner country I is the ratio of aggregate bilateral
exports from country k to the partner I, a
kl, divided by the sum of country k’s exports
to all partner countries. Although the latter weights vary over time we use time invariant
measures α
kl by averaging monthly data over the period 1993:01 to 1998:01. Then, weights
w
kl are determined as

akl
wkl =         ,

2_ak akl


Wk^=kt = 1, I =1,... 19, k = 1,... 15, I = k .

I

Taking the perspective of country k , our measure of foreign economic activity is accordingly

iPkt =     wkl ln (iPlt) , l = k,

l

with iplt denoting industrial production of partner I. Similarly real effective FX rates are
computed as

skltwplt

ekt => ywki ln -------- , I = k,

l           wpkt

where sklt is the price of country k’s currency in terms of country l’s currency. As in Baum
et al.
9 bilateral nominal FX rates are deflated by means of wholesale price indices.

A.2 Data sources and further comments

Sectoral trade flows and industrial production data are from the OECD databases ”Monthly
Statistics of International Trade” (March 2001) and ”Main Economic Indicators”, respec-
18



More intriguing information

1. ENERGY-RELATED INPUT DEMAND BY CROP PRODUCERS
2. Financial Market Volatility and Primary Placements
3. Analyzing the Agricultural Trade Impacts of the Canada-Chile Free Trade Agreement
4. A COMPARATIVE STUDY OF ALTERNATIVE ECONOMETRIC PACKAGES: AN APPLICATION TO ITALIAN DEPOSIT INTEREST RATES
5. A multistate demographic model for firms in the province of Gelderland
6. SOME ISSUES IN LAND TENURE, OWNERSHIP AND CONTROL IN DISPERSED VS. CONCENTRATED AGRICULTURE
7. The name is absent
8. Impact of Ethanol Production on U.S. and Regional Gasoline Prices and On the Profitability of U.S. Oil Refinery Industry
9. WP 48 - Population ageing in the Netherlands: Demographic and financial arguments for a balanced approach
10. Mean Variance Optimization of Non-Linear Systems and Worst-case Analysis