I wanted to know how you calculate the average volatility when there are missing data in the last 2 hours? Do you average with the available data or do you put zeros instead.
I ask myself this question because of one particular example:
X_volatility = [0.370653952007 0.0 0.0 None None None 0.0 None None None None None None
0.745743922056 None 0.0 None None None None None None 0.0 None
0.181105592323 0.0 0.0 None None None None None None None None None None
None None 0.0 None 0.0 0.0 0.0 None 0.0 None None 0.320611961946 0.0 None
0.0 None None]
Y = 0.0001324
Is the very low volatility to predict due to missing data or just due to bad luck?
A None/NaN value for a given 5-minute slice means that there was no complete price information over this period. This happens for instance when the trading for a stock stops for some reason (large stock movements, etc.) or when there is a problem with the stock exchange data. Therefore, such values are different from 0 values and should not be converted into 0 values.
The 0 values in the table are real zeroes (0 volatility and 0 price change).
The volatility measures the price movement amplitude of the stock over a period. Thus, a low volatility over the last two hours of the trading day just means that the stock price did not move much over this period. There are many ways of defining the volatility. It is in any case calculated across the whole provided data set in a consistent manner.