This is calculating volatility based on the log of the returns. I am running some Monte Carlo simulations for which I need returns/volatility and I do not think this is useful for my purposes… Any suggestions? Any simple explanation why they do this (found plenty of explanations that require a math degree)
Log return volatility in practice should be almost the same as percentage return volatility. In small percentage moves, they are almost indistinguishable. Unless you are doing something very sensitive that needs percentage returns specifically, log returns should be fine. In some applications, like derivatives pricing, log returns is preferred, since it is mathematically compatible with many close form formulas. However, if you only want estimates and don’t care about exact mathematical results, it shouldn’t matter.
Anyway, if you really want a specific type of volatility measure, why don’t you just use BDH to pull a time series and calculate it yourself?
ohai i remember logging shit when i was in college because people told me thats the way to do it. but now i feel like percentage changes make more logical sense. what the hell is log anyways?
Log returns are not that difficult to comprehend when you think of the subject in terms of something you are familiar with. Like World of Warcraft. In WOW, the original level cap was 60. Numerous expansion packs were introduced throughout the years with the level cap increasing with each new expansion pack. The current level cap it 110, I think. The amount of experience points required to achieve and rewards given at each new level cap increased exponentially, similar to continuously compounded returns. So, if one were to compare levels on an equal scale (for example something at level 60 vs something at level 90), one would have to take the log metric of whatever was being measured. Then, an equal comparison of this or that could be made.