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



By assumption the error terms in (12) have conditional zero mean and finite variance,
E[β
kt|v] = 0, Var[ektv] = Z2(v) < .                                                     (13)

Given the nonparametric nature of the regression model in (12) on the one hand and recalling
on the other hand that both variables y
kt and vkt are obtained from linear projections the
model in (12) actually formalizes a semiparametric approach. We evaluate the (unknown)
conditional mean, a
k(v), using the locally linear estimator32,33 which is the first component
of a
k = (ak0) , ak1))that solves the minimization problem

T

min Q(v) = min ^" K f-—kt-k~^)
ak                 (0) (1)                   h

ak ,ak t=1


[ykt - a(k0) - a(k1) (v - vkt)]2.


(14)


K(.) and h are a symmetric kernel function and the bandwidth parameter, respectively.
Obviously ^
k(v) solves locally a common least squares problem. Weights associated to sample
values y
kt depend on the distance between vkt and v , the bandwidth h and the employed
kernel function. For our purposes locally linear estimation is implemented by means of the
Gaussian kernel

(-2 u2) ∙

A particular problem in semiparametric regression is to select the bandwidth parameter
h.
34 Owing to the large number of empirical models employed for estimation and recursive

forecasting, a data driven bandwidth selection is infeasible to implement for this empirical
study. Therefore a common rule of thumb bandwidth choice is preferred, namely
where σ
v is the empirical standard deviation of vkt and T is the sample size.

h = σv


(⅛) -


(15)


To illustrate the precision of semiparametric estimates pointwise confidence bands for
ak (v) could be obtained from quantiles of the Gaussian distribution and some variance esti-
mate Z
2(v) or from resampling techniques as outlined in Neumann and Kreiss.35,36 The latter,
applied in this study, have the advantage to account for the potential of heteroskedastic error
terms where the particular form of heteroskedasticity is left unspecified.

3.4 Results

Having introduced the basic tools employed to infer on the relationship between FX uncer-
tainty and trade growth, the partial linear model (11) and the semiparametric specification
(12), we will now provide some model diagnostics for the former and selected estimation
results for both models applied. Since we have investigated 150 export and 140 import equa-
tions we refrain from reporting test statistics in detail but will rather provide test decisions
on an aggregated level.

insert Table IV about here



More intriguing information

1. Categorial Grammar and Discourse
2. he Virtual Playground: an Educational Virtual Reality Environment for Evaluating Interactivity and Conceptual Learning
3. The name is absent
4. Ultrametric Distance in Syntax
5. Dual Inflation Under the Currency Board: The Challenges of Bulgarian EU Accession
6. Transfer from primary school to secondary school
7. The demand for urban transport: An application of discrete choice model for Cadiz
8. Regional dynamics in mountain areas and the need for integrated policies
9. Towards Learning Affective Body Gesture
10. The name is absent
11. Fortschritte bei der Exportorientierung von Dienstleistungsunternehmen
12. Education Research Gender, Education and Development - A Partially Annotated and Selective Bibliography
13. Does South Africa Have the Potential and Capacity to Grow at 7 Per Cent?: A Labour Market Perspective
14. Evaluating Consumer Usage of Nutritional Labeling: The Influence of Socio-Economic Characteristics
15. DEMAND FOR MEAT AND FISH PRODUCTS IN KOREA
16. Evaluating the Impact of Health Programmes
17. Program Semantics and Classical Logic
18. The name is absent
19. Large Scale Studies in den deutschen Sozialwissenschaften:Stand und Perspektiven. Bericht über einen Workshop der Deutschen Forschungsgemeinschaft
20. Institutions, Social Norms, and Bargaining Power: An Analysis of Individual Leisure Time in Couple Households