Getting back to risk measurements you can’t ignore VaR: it’s a well-known and established risk statistics, sanctified by watchdogs all over the world, that risk managers look at in an attempt to predict the potential downside in a financial asset or portfolio of assets exposure. There are many approaches to VaR computations that risk managers use to solve the problem at hand (i.e. measure risk) and sometimes provide top executives with many a version of it: namely historical VaR and parametric (or Gaussian) VaR.
The chart above shows returns on an index fund (the Vanguard REIT Index) coupled with two VaR measurements, one is historical and the other is Gaussian. Depending on which you consider your risk picture might be somewhat more, or less, optimistic. But that is normal: you will never find a one-of-a-kind better risk model, you will always have to use more than one to get to grips with risk management. Choose the one you want, but don’t you use it until 5 digits after the decimal point. It might be risky, after all.