Modelling asymmetric exchange rate dependence
read more
Citations
Variational Mode Decomposition
Measuring financial contagion: A Copula approach
Copula Modeling: An Introduction for Practitioners
Generalized autoregressive score models with applications
Asset Market Linkages in Crisis Periods
References
An introduction to the bootstrap
Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation
A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix
Statistical Methods for Research Workers
An Introduction to Copulas
Related Papers (5)
Frequently Asked Questions (10)
Q2. What is the alternative to a parametric model?
A useful parametric alternative to copula-based multivariate models is a multivariate regime switching model; see Ang and Bekaert (2002) for example.
Q3. What is the function equivalent of (3)?
The density function equivalent of (3) is useful for maximum likelihood estimation, and is easily obtained provided that FX|Y and FY|W are differentiable and FXY|W and C are twice differentiable.
Q4. What is the conditional copula of (X, Y)?
The conditional copula of (X, Y) | W = w, where X | W = w ∼ F X|W(· | w) and Y | W = w ∼FY|W(· | w), is the conditional joint distribution function of U ≡ F X|W(X | w) and V ≡ FY|W(Y | w) given W = w.
Q5. What is the reason why the DM and yen are more dependent on each other?
If the competitiveness preference dominates the price stability preference, the authors would expect the DM and yen to be more dependent during depreciations against the dollar than during appreciations.
Q6. What is the conditional distribution of X, Y, W?
Let the joint distribution of (X, Y, W) be FXYW , denote the conditional distribution of (X, Y) given W, as FXY|W , and let the conditional marginal distributions of X | W and Y | W be denoted FX|W and FY|W , respectively.
Q7. What is the conditional distribution of X, Y, and W2?
Failure to use the same conditioning variable for F X|W, FY|W, and C will, in general, lead to a failure of the function FXY|W to satisfy the conditions for it to be a joint conditional distribution function.
Q8. What is the simplest way to extend Sklar’s theorem?
With a corollary to Sklar’s theorem, given in Nelsen (1999) for example, the set of possible parametric multivariate distributions increases even further, as the authors are able to extract the copula from any given multivariate distribution and use it independently of the marginal distributions of the original distribution.
Q9. What is the conditional bivariate distribution of (X, Y, W)?
Then the conditional bivariate distribution of (X, Y) | W can be derived from the unconditional joint distribution of (X, Y, W) as follows:FXY|W(x, y | w) = fw(w)−1 · ∂ FXYW(x, y, w) ∂w , for w ∈
Q10. What is the test of the normality of the distribution of each exchange rate?
The Jarque–Bera test of the normality of the unconditional distribution of each exchange rate strongly rejects unconditional normality in both periods.