Portfolio VaR by Var-covar i.e. Parametric method

Discussion on Value-at-Risk Models
Forum rules
DISCLAIMER: We do not warrant or represent that this forum or its content is free of viruses, worms or other code that might be contaminating or destructive. We cannot guarantee that documents or files downloaded from the Site will be free from viruses and we do not accept any responsibility for any damage or loss caused by any virus. Accordingly, for your own protection, you must use virus-checking software when using the forum. You must not post or provide to us via the forum, any document or file which you believe may contain a virus. You must virus check any document or file which you intend to post or provide to us via the forum. You must ensure that any document or file you intend to post to the forum does not contravene any applicable laws or contravene any person's legal rights. We do not accept any responsibility for any damage or loss you may suffer.
Katherine Gobin
Posts: 1
Joined: Wed May 15, 2013 8:20 am

Portfolio VaR by Var-covar i.e. Parametric method

Postby Katherine Gobin » Wed May 15, 2013 12:03 pm

Dear Carol madam,

Last week only I had bought the Volume IV - Value at Risk models and then I had realized I need to buy the first 3 volumes also in order to have the structured and clear understanding of Market Risk. I am expecting to get the delivery of rest three volumes in couple of days.

Madam, I am new to market risk and currently in my office, I am a part of project team working on Market Risk. I understand I need to go through all the volumes soon and eagerly waiting for remaining volumes to arrive which undoubtedly will be treasured asset for me.

In the mean time, madam I shall be highly obliged if you can guide me to understand the methodology when it comes to a portfolio. I am bit confused about the Var covar methodology as I have encountered different formulas when it comes to Var covar method.

Suppose I have a portfolio with two equities, two forward Forex transactions and two bonds. For equities, I have some market index as risk factor, when it comes to Forex transactions, (assuming I am dealing with GBP, EURO and USD with transactions maturing in say 91 days), my risk factors are GBP 3 month LIBOR, EURO 3 month LIBOR, USD 3 month LIBOR and respective exchange rates.

Similarly, I have bonds cash flows and using the interpolated discount rates, I have mapped the cashflows into respective riskfactors.

My approach :

(1) I construct a single table where I have all the risk factor rates. Then, I calculate returns of all risk factor rates. Take mean and calculate the covar of returns.

(2) Calculate the portfolio value across all transactions using the current rates.

(3) VaR = (Z value) * portfolio value * stdev(portfolio returns)

Somehow, I think (and I am surely stupid to think this way) this is not the correct way to do it.

Or I calculate the individual VaRs for equity, forex and bond transactions and use the formula

portfolio VaR = (eq_weight * eq_VaR)^2 + (forex_weight * forex_VaR)^2 + (bond_weight* bond_VaR)^2 + 2 eq_weight*forex_weight * correlation(eq, forex) * eq_VaR * forex_VaR .... and so on.

Madam, I am sincerely apologize for raising such a stupid question and wasting your valuable time, but if can vaguely give some guidance as to when it comes to portfolio VaR, it will help immensely to novice like us. I do understand I need to read the literature and I am certainly taking the efforts, but owing to my non - Statistics background, I am finding it bit difficult to understand thing easily.

I also apologize to you for writing such a long mail.



Posts: 1
Joined: Fri Jan 24, 2014 4:54 am

Re: Portfolio VaR by Var-covar i.e. Parametric method

Postby Toby_7 » Sat Feb 01, 2014 11:58 am

Thanks for the post!

Return to “Volume IV”

Who is online

Users browsing this forum: No registered users and 1 guest