Calculating 20-Day Annualized Variance from Daily Stock Returns
Summary
The document presents a SQL approach to estimating annualized variance for Chinese stocks. It first calculates daily close-to-close returns for each instrument, then applies a rolling 20-observation standard deviation, squares that value, and multiplies by 250 to annualize it. The output is organized by instrument and date, making the resulting series suitable for further strategy research.
The post is framed as a question about using the calculation inside a strategy, but it does not explain how to connect the query to a trading system or validate the feature. It also does not discuss assumptions behind the annualization factor, missing data, the rolling-window behavior, or whether the intended measure should use sample or population standard deviation. Treat the query as an example calculation rather than a complete implementation guide or evidence of a profitable signal.
Key ideas
- Daily close-to-close returns provide the input series for the variance calculation.
- A rolling 20-day standard deviation is squared to produce a variance estimate.
- Multiplying the daily variance by 250 expresses it on an annualized trading-day basis.
- The post does not show how to integrate the query into a live or backtested strategy.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.