I have read the comments on this thread. Surely the point of this forum is to try to help each other ? After all, we are the small guys, and we need each other.
With this in mind, here are the results of some research, which may or not be useful to participants of this thread.
I looked back at Hourly data for GBP (spot) for the period from 1/7/03 to 19/8/04, there were 295 trading days in my data set. For each day, which in this case is defined as 24h00:00 to 23h59:59 GMT, I looked for the hour bar with the highest tick volume (couldn't use volume as it's spot hence no volume available, tick data is next best), and then at the closing price for the day, relative to the prices during the highest volume bar.
OK, I know, before someone starts shouting, this may be useless in practise for actual trading purposes, because how could we know at the time whether we had just seen the hour bar that would eventually have highest volume for the day ? However, the point of the exercise was to see whether there was any sort of bounce occurring as a result of the high volume, or whether the results would appear to be random. If the latter, then we might conclude that (tick) volume studies, for hour bars and GBP at least, are essentially a waste of time. I will leave you all to draw your own conclusions from this small data set :
No of days = 295
No of days closing (at 23h59:59) WITHIN highest volume hour bar price range = 132
No of days closing (at 23h59:59) ABOVE highest volume hour bar price range = 92
No of days closing (at 23h59:59) BELOW highest volume hour bar price range = 71
(HVB = high volume bar)
HVB was an UP bar closing INSIDE the HVB price range = 66
HVB was an UP bar closing ABOVE the HVB price range = 57
HVB was an UP bar closing BELOW the HVB price range = 24
HVB was a DOWN bar closing INSIDE the HVB price range = 66
HVB was a DOWN bar closing ABOVE the HVB price range = 35
HVB was a DOWN bar closing BELOW the HVB price range = 47
So, for the UP bars, 57 out of 147 times the price finished up at the end of the day above the high of the bar. For the DOWN bars, 47 times out of 148 times the price carried on falling below the low of the HVB. IOW, 191 times out of 295, the high volume bar seemed to create a price reversal, which (to me) appears to be more than just a statistical aberration. Or maybe not, who knows.
Whether you feel that this is tradable is another matter, but I find the results interesting nevertheless. The next step might be to look at futures data, and the closing ranges on these HVBs.
rog1111
