Volatility Estimators - AlgoJi

Volatility estimators are metrics which try to measure/reflect the volatility changes. The calculations on advanced estimators below can be found from literature by the respective authors.

Coefficient of Variation (%CV)

%CV = (Standard deviation / Mean) x 100%

The above formula simply means that in order to compute %CV, you simply measure the standard deviation and mean of stocks data and then divide standard deviation over the mean. Finally to have the result in percent, you will need to multiply it by 100%.

1.) When you are about to invest stocks, it is suggested to measure the volatility in terms of %CV. Compare the risk of those stocks according to their volatility and then invest those stocks with low volatility. These types of stocks are low risks, more predictable and stable.

2.) The recommended number of samples in the measurement is more than 36. If you have historical samples of around one year before; the more accurate will be the result.

Close-to-Close Estimator

Pro
It has well-understood sampling properties.
It is easy to correct bias.
It is easy to convert to a form involving typical daily moves.
Con
It is a very inefficient use of data and converges very slowly.

Parkinson Estimator

Pro
Using daily range seems sensible and provides completely separate information from using time-based sampling such as closing prices.
Con
It is really only appropriate for measuring the volatility of a GBM process. In particular it cannot handle trends and jumps.
It systematically underestimates volatility.

Garman-Klass Estimator

Pro
It is up to eight times more efficient than close-to-close estimator.
It makes the best use of the commonly available price information.
Con
It is even more biased than the Parkinson estimator.

Rogers-Satchell Estimator

Pro
It allows for the presence of trends.
Con
It still cannot deal with jumps.

Free Webinar Series on Algorithmic Trading

29th May | From Concept to Automation – 9 Trading Strategies
30th May | Algorithmic Trading using Machine Learning in Python
31st May | Kickstart Guide to Algorithmic Trading
1st June | Algorithmic Trading Using Amibroker and MTv3