# How Measure Rate of Variance In Volume Traded Over Consistent Time Periods

on
Is this Statistical Question as simple as i hope it is...

If I have a 5 Minute Time Bar for measuring trade volume within a Stock Symbol (any symbol)

Which is 300 seconds in total (5 x 60 seconds in a minute = 300)

and i divide the 300 seconds into 10 equal segments of time -- 30 seconds each

and i measure the buying volume that took place from trades during EACH of the 10 segments --- HOW DO I MEASURE THE RATE OF BUYING PER SEGMENT IN RELATION TO THE 10 SEGMENTS to quantify how smooth was the trading done... Measure the Variance ? How do i measure the consistency of the buying ?

For example:

If each of the 10 segments of 30 seconds had 1000 shares traded then the rate of trade volume consistency of trading per each segment would be exact or the same rate flow of volume

1000  - 1st segment
1000
1000
1000
1000
1000
1000
1000
1000
1000 - 10th segment

But in the real world you get more inconsistent flow something like this... and i would like to understand how to measure the variance here ?

1250 - 1st segment
800
900
1600
550
1700
1250
1350
1100
1500 - 10th segment
Comment
Watch Question

Do more with

EXPERT OFFICE® is a registered trademark of EXPERTS EXCHANGE®
HelpDesk Technician
Commented:
Hi Enflow,

We can measure the variance by using the Sample Variance equation, although usually this number is difficult to apply meaning to. Instead, usually we take its square root to find the Standard Deviation which is usually a much more useful number. The Standard Deviation is a number used to tell how measurements for a group are spread out from the mean, or the expected value.

I've included the equation and how to find both the Sample Variance as well as the Standard Deviation below. I calculated all of the desired values using your data set, you can find those here as well.

Please note that if you have the entire population data, and not a sample, that you will want to use 'n' instead of 'n - 1' to find the population variance

Commented:
@Michael,

Hi... That is very interesting... thanks for your time...

Please note that if you have the entire population data, and not a sample, that you will want to use 'n' instead of 'n - 1' to find the population variance

Sooo. I think i would be using n instead of n-1 since all the data i gave you in my sample is all the data of that 5 minute bar which is a separate data set from the NEXT 5 minute bar with its own data...

I sent you a message thru EExch... please email me back... thanks... CJ
HelpDesk Technician

Commented:
You're welcome!

I went ahead and did the Population Variance and S.D. instead of the Sample Variance and sample S.D. here. As you can see, it's a pretty similar answer.

Cheers
PopulationVariance.PNG

Do more with