Standard Deviation and Statisticians

[kleinfelter]

I mentioned that I was using Inflation Average of 3% and Inflation Standard Deviation of 1% to a statistician, and he reminded me that SD is the square root of the Variance, and is typically a number like 1.5, not a percent.

When I look at "Inflation - Std Dev" and "Return - Std Dev", I see percentages.

What is expected in these fields? Are they expecting the Coefficient of Variation? Or perhaps something with an informal definition that a statistician would not recognize? Or maybe I should just treat "%" as "times 100" and if I'm using a standard deviation of 1.2 I should enter 120%?

I'm not trying to quibble with the label. I'm trying to figure out when I say "Inflation - Std Dev is 1%" what is it that is 1% (or 1% of what)?

[kleinfelter]

Ah. I found the post with the Excel VBA. That reveals that an "Inflation - Std Dev" of x% means that the value of one standard deviation is x%. An example makes it clearer.

If I choose stdDev of 1% and avg of 3%, I'll see for the series of years:

• Inflation centered on 3%
• 68% of years will be 2-4% (1 standard deviation = inflation rate changes by 1%)
• 95% of years will be 1-5% (2 standard deviation = inflation rate changes by 2%)
• 99% of years will be 0-6% (3 standard deviation = inflation rate changes by 3%)

"Inflation - Average" sets the midpoint. "Inflation - Std Dev" sets the *amount* by which inflation will vary for the most common 68% of samples.