Scenario-Based Attribution of Correlation Risk in a t-Copula Portfolio
Summary
The document asks how to measure portfolio sensitivity to dependence in a multivariate t-copula model, in a setting involving electricity futures and rank correlations estimated with Kendall’s tau-b. The proposed finite-difference calculation perturbs the correlation matrix, but the answer recommends a scenario comparison instead: run the model with the empirical dependence matrix, then run it with a zero matrix, and treat the difference in portfolio outcomes as an attribution of correlation risk.
The author describes using empirical return distributions and simulated copula draws to generate portfolio outcomes. They report that their VaR simulations align with desk experience across ten books and roughly 500 trading days, but this is not evidence validating the correlation attribution itself. The zero-matrix comparison is a broad counterfactual that removes modeled dependence; it does not isolate the effect of individual correlations or provide a local sensitivity. Results also depend on the chosen marginal distributions, copula, and matrix construction, and the text gives no detailed implementation or validation procedure.
Key ideas
- A finite-difference perturbation of a correlation matrix is one proposed way to approximate portfolio sensitivity.
- The answer instead compares portfolio outcomes under empirical dependence and a zero correlation matrix.
- The difference between those runs is presented as an attribution of correlation risk.
- Kendall’s tau-b is selected to estimate rank dependence in the presence of ties.
- The proposed comparison gives an aggregate counterfactual, not a decomposition by individual correlation entries.
Tags
Full text
# Answer by CasusBelli (score 1)
# Correlation sensitivity in multivariate $t$-copula for portfolio VaR of electricity futures using Kendall's tau-$b$ correlation matrix
My t-copula model captures the daily dollar returns of a portfolio of approximately 400 assets. I am curious if there's a generally accepted way to quantify the sensitivity of portfolio movements with respect to the underlying correlation matrix. My first instinct is to try a discrete approximation, such that If C is my correlation matrix, and X is my current returns:
$$\frac{dX}{dC} \sim [X(C + 0.0001) - X(C - 0.0001)] / 0.0002$$
Is this a valid approach? Your help is greatly appreciated!
EDIT: Forgot the outer parentheses on the numerator.
### Further Edits from own comments
- I applied a univariate t CDF with 3df to a multivariate t distribution in Python with 3df. Then, having gotten values on (0, 1) I applied the respective inverse probability transform for each of the data to rescale to the original level. I then applied those returns to the previous closing prices and multiplied by the notional tied to the series, and summed the result. My idea in the above was to quantify correlation risk. I chose 0.01% somewhat arbitrarily, but the idea is: how can I perform a correlation / dependency risk attribution?
- I’m just interested in separating correlation risk from price risk and (as I’m dealing with electricity futures) generation / volume risk
- My question was specifically about correlation risk sensitivity and attribution. I didn’t see a reason to get into VaR. But, since we’re now into that rabbit hole, my VaR simulations are in line with data we’ve seen from our trading desks over the past 500 trading days and across ten books.
- As for correlation selection: copulas feature nonlinear transformations, so a rank correlation matrix is necessary, and Kendall's tau handles ties (which I have) better than Spearman does.
- I’m not optimizing anything. I’m just modeling a VaR for a portfolio of correlated assets. I know from the literature that these are best described using a multivariate t copula with v=3. I estimate the empirical probability distributions of my observed returns. I get my rank correlation matrix. I want to measure my correlation risk and correlation sensitivity.
## Answer by CasusBelli (score 1)
https://quant.stackexchange.com/a/58740
Instead of thinking "at the margin", I've opted to conduct an attribution of sorts, by running the copula with the empirical Kendall's tau-b correlation matrix and again with a zero matrix. The difference between the two scenarios represents correlation risk.Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.