[mister ed]

A Volume moving average; I like the idea so I know if the volume in the market is higher than average, indicating outside money flowing in/out and good potential for a trend move, or lower than average indicating the locals might spend the day playing with stops and we have a range sort of day. That’s the (simplified) theory, but the practical side of it is that volume varies depending on time of day so using a normal moving average calculation (take the last “x" periods, add them up, divide by “x�) will not give a reliable indication of what the average volume is for a particular time of day.

Better to take the volume for a certain period of time, say the 15 minutes between 8.30 and 8.45 am, and compare the volume during this period to the volume between 8.30 and 8.45 am for previous days. Maybe add up the volume for each of the 8.30 to 8.45 am time slots for the previous 20 days, and divide by 20 (20 is not carved on a stone tablet, use whatever number of periods suits; also this calculation is for a simple MA, could also use an exponential, weighted, or whatever suits). Do this for each 15 minute period during the trading session, and at a glance we can tell if the volume is above or below average for this time of day.

I suppose plenty of software packages will do this sort of thing, I know Ensign does it, but I don’t have Ensign. So what I did was get the volume data for each 15 minute period of the ES from 8.30 to 3.15 (Chicago time), put it in Excel (if you don’t have Excel, you could use the Open office spreadsheet software – free download) and did the calculations, this is what I came up with for the average volumes (simple MA) every 15 minutes using the data for the dates between Sep. 19 and Oct. 17, 2007 inclusive:

Chicago time: Average volume
(15 mins ending…) (no. of contracts traded)
8.45 86,985
9.00 74,498
9.15 83,731
9.30 64,551
9.45 67,687
10.00 56,384
10.15 57,069
10.30 52,050
10.45 47,209
11.00 36,678
11.15 31,807
11.30 24,842
11.45 25,120
12.00 22,855
12.15 28,098
12.30 35,421
12.45 39,401
1.00 36,565
1.15 49,971
1.30 51,979
1.45 52,430
2.00 47,979
2.15 60,035
2.30 58,578
2.45 58,518
3.00 79,956
3.15 67,798



I don’t know how often I will have to update these figures, if they change dramatically each day then there would not seem to be much point to them, but I doubt they do. I might update them each day for a couple of weeks and see if there are any big swings in values; if, as I expect, there is not, then I will probably update them once a week, or even extend the period out to 30, 40, or more, days – it’s a simple matter of just adding more data.

I have attached the spreadsheet I designed (ummm…. designed is too fancy a word for what I cobbled together) to do the calculations, for those who may find it useful.

Any thoughts, suggestions, modifications ?

es vol data.xls

[januson]

Interesting theory, before I code it I would like to hear something more about the idea.

As I understand you assume that the volume between 0830 and 0845 from the previous day should be compared with current day. Should a moving average then be calculated backwards 20 days between 0830 and 0845 and then be applied to current day 0830-0845? Same procedure follows with other times!

[darthtrader2.0]

I'm been thinking about adding something like this to what I watch but just for the opening range to try to watch out for trend days.
I think if you do it for the whole day, a broader view of things might have more utility because I'm not exactly sure what information you are gaining from knowing that the current volume from 10:45-11:00am is higher than some windowed average of prior days at that exact time.
Some good data points to me would be
8:30 - 9:30
8:30 - 10:00
8:30 - 11:00

After that to me your getting into a higher/lower overall volume day. If a genie looked into the future and told you monday would have higher/lower volume overall on the day than the average of the last 20 days...well I really don't see how that information would be of much use.

[januson]

hehe.. couldn't agree more, however people are watching those moving averages as an overall for n-bars back in time.

I thnink I understand now, but
8:30 - 9:30
8:30 - 10:00 <- shouldn't it be 0930-1000
8:30 - 11:00 <- and 1000-1100

I cannot understand why we should accumulate from 0830? Shouldn't each interval be compared with the same interval each day?... Ofcause it can be done your way, could be interesting to analyze.

We should also use 13-14, 14-15 and compare with current daytime ?