Annualized Sharpe Ratio Calculation

[jakewillms17]

Wondering why the annualized sharpe ratio calculation doesn't line up with the traditional methodology of annualized return / annualized volatility?

qs.stats.cagr(returns)/qs.stats.volatility(returns)
0.6317448862990539
qs.stats.sharpe(returns)
0.8451565697277529

returns is a daily returns object. All other resources I've worked with give a number like the 0.6317 value (R, online resources).

[jakewillms17]

following up on this, most of the difference was as a result of the 365 days for a period. This should be 252. the rest, however, would have to do with using arithmetic vs. geometric return when calculating Sharpe I believe?

[cantupaz]

I was looking at some results from these two functions. I think the problem is that cagr() and sharpe() use the periods parameter to mean different things. In cagr() the periods parameter is used to calculate the number of years between the first and last return (so it's really meant to be 365). But in sharpe() the periods parameter is meant to be the number of periods in a year, so if returns has a monthly frequency we should use 12, 52 for weekly returns and 252 (or 365) for daily.