CFA certification is professional certifcation by the CFA institute primarily for those with financial background and relevantly related to investment analysis, portfolio management, marketplace analysis skill etc.CFA Program is a 3 level exam series i.e. Level I, II, and III which need to be cleared sequentially. It is being conducted by the CFA institute formerly known as AIMR. Along with clearing the complete CFA Program the candidate needs to  adhere to certain rules and regulations stated by the CFA institute in order to become a CFA Charter holder. You can easily get a link to these norms from here.

The result of the recently conducted CFA June Exam series is out and quite a good percentage of candidates cleared the CFA exam.
Not many across the world have gained a credential of CFA Charter holder. CFA charter holders are well acknowledged across the globe and have left their marks in almost all the possible financial sectors spread worldwide. You name a renowned employer say it be Deutsche Bank, HSBC, JP Morgan Chase and you will find a CFA charter holder working with them; this is their reach. Even the candidates with partial CFA program cleared demonstrate a significant caliber and potential skills over the others. Get a stat result of the CFA charter holders' performance across the world from here.

If you wish to prepare for the next series of the CFA exam or wish to have an overview about the CFA course and content please follow up here.

Heteroskedasticity happens once the variance of the disturbance isn't constant, that is commonly a drag encountered in cross sectional data. Heteroskedasticity doesn't have an effect on the parameter estimates: the coefficients ought to be unbiased, but it will, however, bias the variance of the calculable parameters.

By using R, Heteroskedasticity can be easily removed from our model. “sandwich” and “lmtest” are R’s packages that are required for removing Heteroskedasticity from our model. Install these packages in R and go through their manuals to get a proper and complete working of these modules.

Now, further we construct a variance - covariance matrix using vcovHC() function to find out whether heteroskedasticity exits or not in our data.
>vcovHC(FitLinReg, omega = NULL, type = "HC4")  The presence of variation in the diagonal elements of the above generated matrix concludes the presence of heteroskedasticity.

Finally use the coeftest() fucntion in R to remove the present heteroskedasticity from the data.
>coeftest(FitLinReg, df = Inf, vcovHC(FitLinReg, omega = NULL, type = "HC4"))

Here we take df equal to infinity because of the presence of large number of variables. df stands for degree of freedom.

Also learn in detail to remove heteroskedasticity from your model using R and also using Excel.

Find an answer to what actually is heteroskedasticity????

As we have seen that quite a lot of portion of FRM Exam is based on numerical calculations and some of those questions are tough as well. We thought that it is best to find a set of questions, that are purely formula based (And if you know the formula, you can easily do them) and how many of them require deeper understanding of material and how many require combinations of formulas and concepts (maybe interpreting the data, and finding the relevant values to be used and then combining 2-3 formulas to get to the result).

If we look at the FRM Exam Practice Paper 2009, we find that 50% of the questions are purely formula based questions. This is good news!

If you are able to understand the question, remember few formulae, use the calculator properly and make no silly mistakes, you can crack a majority of the numerical questions as well!

Within the formula based questions for Value at Risk, the most common question (usually multiple number of questions are based on this concept) is related to finding the portfolio VaR. Calculating the portfolio VaR requires the calculation of portfolio volatility, for which we have the simple formula:

VaR(Portfolio)= wa^2*σa^2 + wb^2*σb^2 + 2*wa*wb*σa*σb*Corelation(a,b)

Similarly another common formula is calculating the annual VaR from daily VaR (or changing the time duration of VaR) or calculating the significance intervals. These again use simple formulas:

VaR (for time period T) = VaR(Daily) * Sqrt(T)
VaR (Significance Level alpha) = Z (Alpha) * Sigma

Similarly Questions on Quantitative Analysis, which are formula based, are based on EWMA, GARCH formulas, which are quite simple to calculate, if you know the formula.

Typical formulas would be:

Return = ln(P(t)/P(t-1))
(Sigma(n))^2 = lambda * (Sigma(n-1))^2 + (1 – lambda) * U(n-1)^2

Similarly the formula based questions from the rest of the parts of the FRM exam are not very difficult. If you are able to remember them and apply them in the FRM Exam, your chances of success would be very high.

EduPristine realized this long ago that covering full Risk Management Syllabus on the last day and remembering a whole lot of formulas can be a daunting task. Pristine summarizes a whole lot of important concepts, formulas and questions based on those formulas in VisualizeFRM®.


You can also refer to GARP website for FRM exam updates.