A lot of people adamantly feel that there is a strong relationship between price and volume and this hypothesis makes sense. However, I've never observed this, casually looking at charts and I wonder if there has been any empirical evidence presented to support this hypothesis. For example, in the attached 5 minute chart for ES last week, I highlighted the peaks in average volume and they do not appear to display any pattern that is discernable. The peak shows up when the price is peaking and bottoming, which you would expect, but also when it's moving up or down near points of resistance. In addition, I analyzed ES quotes for the past 3 months and here is what I found. I divided the data into four quartiles of volume and it produced the following means: Avg Vol. Avg Close Avg C-O 1816 1426.40 2.9% 5531 1424.65 3.2% 10423 1421.54 2.7% 34908 1419.50 -1.6% From this it would appear that when the volume is greatest, the price tends to drop on that bar and shows up when the prices are lowest, contrary to some of the opinions expressed on ET. I realize this is a limited sample, yada, yada, yada. But my question is has anybody presented any data to support the relationship between price and volume, or is this something that just "feels right?"
Woe betide you, for the Jacobeans shall descend on thee with the totally Aristotelian theory that volume leads price. Were they but to observe P and V at the tick level, they would repent, but apparently they cannot follow the tape that fast.
Here is the 15min ES chart from Friday. Do you see anything on this chart that told you a change in trend was coming long before it bottomed? <img src="http://www.elitetrader.com/vb/attachment.php?s=&postid=1342127">
BB, when I look at that chart, I see a spike occurring after the price had already fallen about as far as it was going to fall. If you look at a 2 min chart for the same period, all of the volume came on the same bar (Low1422.75 ). Then the price drifted for a while before heading north. I guess I don't see the predictive power in your example. What Hype said made more sense to me. I created and hereby claim all rights to the Anchovy Market Theory. At the Monterrey Aquarium, they have a huge glass cylinder containing anchovies, all swimming in the same direction, either clockwise or counterclockwise. Then, for some reason, one of the anchovies decides to swim the opposite way and then sometimes all of the anchovies end up swimming the other way. Sometimes the radical anchovy is ignored. There is no way (for a human being, anyway) to predict when one of the anchovies will change direction. Price is the leading anchovy and volume is all of the other anchovies, who can either go along or not, i.e. volume confirms the move but doesn't predict the move. Anyway, I don't want to start an argument with anybody, I just wanted to know if anybody ever studied this relationship through backtesting or some scientific method. I only recall seeing anectodal evidence in the books I read so far. In my experiments so far I have been unable to make volume work to predict price, but my mind is wide open.
If you cannot see the relationship on that chart theres nothing more I can say. People just dont see things the same way, I guess, and thats what makes all this possible.
Checkout bolter's thread on volume profile: http://elitetrader.com/vb/showthread.php?s=&threadid=63202