How does MC use the inflation rate?
I would appreciate an explanation of how the MC calculation uses the inflation rate that the user sets in Assumptions screen. For example, is the mean of the probability distribution of real returns for each asset class simply increased by the assumed inflation rate? Does the inflation rate affect the variance?
RSS
Yes, the real interest rate is hit by the inflation factor to get nominal interest rates. The inflation rate is not a state variable in the monte carlo so there should be no effect on the variance (Larry will have to check me on that but I'm pretty sure I'm right).
Best,
Dick Munroe
Dick,
Thank you. I noticed that if I run a single asset class in MC, the "projected" return is the sum of the inflation factor and the average real return. I would also like to understand how the inflation factor alters the parameters of the asset class distribution, so that I might be able to better interpret the results.
Finally, are there any references that might help me link inflation factor and nominal return when not running MC?
Regards,
Mike
Yes, for "large" periods of time the conversion from real to nominal is:
(1 + nominal) = (1 + real) * (1 + inflation)
You then subtract 1 to get the %age interest rate.
As the period of time approaches 0, the conversion from real to nominal approaches:
nominal = real + inflation
as a limit (Larry asserts this but hasn't shown me the math that backs it up so I can't provide further details and have forgotten most of my calculus so I can't figure it out off the top of my head).
Best,
Dick Munroe
p.s. I'm sure that all the results in the MC are reported in real terms, so you shouldn't be seeing anything that looks like 1+nominal + real in the results. It's possible there's an issue in the MC (we're chasing a couple of things in the Choleski magic that allows us to deal with covariance of assets but haven't gotten any results to report at the moment, it's a research thing) but we haven't seen anything like that. All bets are off if you aren't running with aggressive spending when running the MC module though. Anything other than aggressive spending skews things way towards safe results (as you would expect).