volume difference between CL G8 and CL G8=1

[clearpicks]

I wrote a script to calculate the volume difference between CL G8 and CL G8=1. I edited the study to set s1 to CL G8 and s2 to CL G8=1 and the script was applied on CL G8 1min chart.

Can someone explain why there are a few bars where the returned volume diff are negative? I think all trades of CL G8=1 should be subset of CL G8, therefore the volume of CL G8 should be always larger than CL G8=1. Do I miss something?

The volume of CL G8 at 9:08 is 76 while that of CL G8=1 is 97, which results in a negative reading of vol(CL G8) - vol(CL G8=1).

- Clearpicks

[clearpicks]

screenshot

[shortandlong]

A few things:

1. The # of ticks for CL G8=2 (pit/day session) is almost negligible (maybe a couple hundred per day).

2. Most days the pit session trade volume for each trade is always zero. This is the case for 12/19, based on Time & Sales.

3. Even though the anomalies on your histogram are after the start of pit trading at 9am EST, it has nothing to do with the pit session since all volume for CL G8=2 is zero.

Based on the above I'd say this is another example of inconsistencies between eSignal's databases/servers or some other data integrity problem.

Diff'ing the Time & Sales for CL G8 and CL G8=1, while ignoring zero volume trades would further pinpoint the problem. Could be missing trades, extraneous trades, trades with wrong volume, etc.

The negative value around 9:40am seems to be associated with a bad price (~90.95) as well. I don't see that spike on my charts.