Three parameters define economic credit capital: the risk horizon, ie the time horizon over which the risk is being assessed; the confidence level, ie the quintile of the loss distribution; and the definition of credit losses, ie whether mark-to-market losses are considered in addition to default-only losses. The probability of default is not a parameter within the control of the risk manager, but an input into the capital calculation process that he has to estimate. Therefore Choice 'c' is the correct answer.

Choice 'a' is the correct answer. The other choices represent non-sensical statements.

CreditRisk+ treats default as a binary event, ignoring downgrade risk, capital structures of individual firms in the portfolio or the causes of default. It uses a single parameter, λ or the mean default rate, and derives credit risk based upon the Poisson distribution. Therefore Choice 'c' is the correct answer.

Remember that VaR is merely a multiple of the portfolio's standard deviation. The multiple is determined by the confidence level, and for a 99% confidence level this multiple is 2.3264 (=-NORMSINV(1%) in Excel). Therefore one standard deviation at this level of confidence would be equal to VaR/2.3264.

In addition to the Z-value at 99% confidence, you should also remember what the Z value is for a 95% level of confidence, as PRM questions may expect you to know these. The standard Windows calculator allowed in the exam does not allow you to calculate these, so it is safer to just remember these values.

Credit VaR is the difference between the expected value of the portfolio and the value of the portfolio at the given confidence level. Therefore the credit VaR is $85m - $ 60m = $25m. Choice 'b' is the correct answer.

Note that economic capital and credit VaR are identical at a risk horizon of one year. Therefore if the question asks for economic capital, the answer would be the same.

[Again, an alternative way to look at this is to consider the explanation given in III.B.6.2.2: Credit Var = Q(L) - EL where Q(L) is the total loss at a given confidence interval, and EL is the expected loss. In this case Q(L) - $100-$60 = $40, and EL = $100-$85=$15. Therefore Credit VaR = $40-$15=$25.]

All product pricing should reflect the cost of maintaining the liquidity required to support a product. This is regardless of the accounting treatment for the product, ie irrespective of whether the product is on or off balance sheet. Therefore statement I is incorrect.

The time horizon to consider for liquidity risk management is determined taking into account a number of factors, which include both the speed at which new sources of liquidity can be generated and any applicable regulatory requirements. Statement II is correct.

Managing collateral, both collateral received and collateral posted with counterparties, is an important aspect of liquidity risk management as liquidity problems often manifest themselves in the form of margin calls requiring collateral to be posted. Statement III is therefore correct.

The maturity period of the different sources of capital funding for a bank, for example equity capital, preferred shares, long term debt etc quite clearly has a significant impact on liquidity needs. Stable sources of funds such as equity or preferred capital, or debt that is not maturing shortly help the liquidity position. Statement IV is therefore correct.

Choice 'b' is the correct answer.

The square root of time rule cannot be applied here because the returns across the periods are not independent. (Recall that the square root of time rule requires returns to be iid, independent and identically distributed.) Here there is a 'autocorrelation' in play, which means one period's returns affect the returns of the other period.

This problem can be solved by combining the variance of the returns from the two consecutive periods in the same way as one would combine the variance of different assets that have a given correlation. In such cases we know that:

Variance (A + B) = Variance(A) + Variance(B) + 2*Correlation*StdDev(A)*StdDev(B).

The standard deviation can be calculated by taking the square root of the variance.

Therefore the combined volatility over the two months will be equal to =SQRT((5%^2) + (5%^2) + 2*0.3*5%*5%) = 8.062%. All other answers are incorrect.

A recap would be in order here: calculating operational risk capital is a multi-step process.

First, we fit curves to estimate the parameters to our chosen distribution types for frequency (eg, Poisson), and severity (eg, lognormal). Note that these curves are fitted at the UoM level - which is the lowest level of granularity at which modeling is carried out. Since there are many UoMs, there are are many frequency and severity distributions. However what we are interested in is the loss distribution for the entire bank from which the 99.9th percentile loss can be calculated. From the multiple frequency and severity distributions we have calculated, this becomes a two step process:

- Step 1: Calculate the aggregate loss distribution for each UoM. Each loss distribution is based upon and underlying frequency and severity distribution.

- Step 2: Combine the multiple loss distributions after considering the dependence between the different UoMs. The 'dependence' recognizes that the various UoMs are not completely independent, ie the loss distributions are not additive, and that there is a sort of diversification benefit in the sense that not all types of losses can occur at once and the joint probabilities of the different losses make the sum less than the sum of the parts.

Step 1 requires simulating a number, say n, of the number of losses that occur in a given year from a frequency distribution. Then n losses are picked from the severity distribution, and the total loss for the year is a summation of these losses. This becomes one data point. This process of simulating the number of losses and then identifying that number of losses is carried out a large number of times to get the aggregate loss distribution for a UoM.

Step 2 requires taking the different loss distributions from Step 1 and combining them considering the dependence between the events. The correlations between the losses are described by a 'copula', and combined together mathematically to get a single loss distribution for the entire bank. This allows the 99.9th percentile loss to be calculated.

The correct definition of CVA is LGD * EPE * PD. All other answers are incorrect.

CVA reflects the adjustment for counterparty default on derivative and other trading book transactions. This reflects the credit charge, that neeeds to be reduced from the expected value of the transaction to determine its true value. It is calculated as a product of the loss given default, the probability of default and the average weighted exposure of future EPEs across the time horizon for the transaction.

The future exposures need to be discounted to the present, and occasionally the equations for CVA will state that explicitly. Similarly, in some more advanced dynamic models the correlation between EPE and PD is also accounted for. The conceptual ideal though remains the same: CVA=LGD*EPE*PD.

If you have not studied PCA prior to the preparing for the PRMIA exams, you will find the handbook to be a bit difficult to understand this topic. However, PRMIA have been asking questions about PCA on their exams so it is important to conceptually know what PCA is (and be able to answer the exam questions which are unlikely to ever require you to do a calculation).

Using PCA, a complex correlated system (eg, the rates in a term structure which are all correlated to each other) is effectively transformed into a 'principal component representation'. A principal component representation expresses each of the independent variables in the system as a linear representation based on 'principal components'. Principal components are derived from the eigenvectors of the correlation matrix. Remember that if you have a positive semi-definite correlation matrix (which will be an n x n square matrix), it will have 'n' eigenvectors, each with an eigenvalue. The eigenvectors will be orthogonal (ie perpendicular) to each other. The eigenvalues for each of the eigenvectors determine which are principal components - in descending order of the eignevalues. Therefore the eigenvector with the highest eigenvalue is the first principal component, the second highest value the second principal component and so on. If there are n variables in the system, then eigenvalue divided by n gives us the proportion of the variation explained by that particular principal component.

In the case of a PCA done on the term structure of interest rates, the first principal component is the 'trend' component, and under a principal component representation its coefficients are given by the first eigenvector. (By the way, remember the next two principal components are tilt and curvature respectively). If these coefficients are the same (they are, for an interest rate term structure), this means all rates move up or down when the value of the first principal component moves. Thust factor weights on the first component show whether there is common trend in the system. Statement I is correct.

The factors to be applied to the principal components are indeed obtained from the eigenvectors of the correlation matrix (and not the covariance matrix), and therefore statement II is correct.

PCA is used to reduce dimensionality, this is true and in fact one of the main reasons why PCA is used at all. Therefore statement III is correct.

Statement IV is false as it is the largest eigenvalues and not the smallest that determine the eigenvectors affecting most of the variation (through the principal components they represent).

Thus Choice 'b' is the correct answer.