Re: How to compute std dev of time series?
beliavsky_at_aol.com
Date: 12/27/04
- Next message: Richard Ulrich: "Re: chi-square test goodness of fit"
- Previous message: Jim Beck: "Re: any successful stories of gambling using those math/stat theorems?"
- In reply to: nomail1983: "How to compute std dev of time series?"
- Messages sorted by: [ date ] [ thread ]
Date: 27 Dec 2004 09:43:49 -0800
nomail1983 wrote:
>1. In general, how do we compute the std dev of a time
>series (of periodic percentage change)?
First fit a model to the time series, differencing it to induce
stationarity if necessary, such that residuals of the model are
uncorrelated and preferably have a constant variance over time. For
financial time series, where daily returns should be uncorrelated if
the market is efficient, one computes log returns as
r(t) = log(p(t)/p(t-1))
where r(t) and p(t) are the return and price for day t. Then the
standard deviation or "volatility" of the time series is just the
standard deviation of the vector of returns. Often the volatility is
annualized by multiplying the daily volatility by the square root of
the number of trading days in a year.
Calculation of "historical volatility" is discussed at
http://www.riskglossary.com/articles/volatility.htm .
- Next message: Richard Ulrich: "Re: chi-square test goodness of fit"
- Previous message: Jim Beck: "Re: any successful stories of gambling using those math/stat theorems?"
- In reply to: nomail1983: "How to compute std dev of time series?"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
|
