The myth of volume as a leading indicator of price direction...

NazSherpa said:
That's a fair question. Volume is known in real time and if you start digging around with these stats you may come to recognize that volume in the first bit of the morning session is a good indicator of how the day will unfold in terms of range. Again, if you're looking for continuation or something to that effect and the volume is not there this might be good to know.
More...
Ok, well that changes the math problem a bit.

Your original statement was general, but I now assume you are only referring to intraday. You seem to be saying that morning volume is correlated to the range in the future (afternoon, etc.).

For someone to test/verify this, they would have to know the time frame in which to measure the morning volume; and the subsequent time frame in which to measure the range.

It seems you would be in the best position to confirm your observation, no? As you know exactly the criteria.
 
NazSherpa said:
NQ has a correlation coefficient of .63 - that's considered a high degree over 20 years. .41 would be moderate, right?
More...
I'd have to defer to something like this:

upload_2021-8-14_18-15-49.png


https://www.dummies.com/education/math/statistics/how-to-interpret-a-correlation-coefficient-r/
 
So we can say that a correlation of .63 is somewhere between moderate and strong, fair? And, that's just looking at daily volume. So, I will repeat my initial observation. If you have something with a high degree of correlation to range and you're looking for continuation - that's one variable that you might want to take into consideration.
 
NazSherpa said:
Watch a chart? Vince, I hate to break it to you but if you read my post- I'm referring to ....
More...
Vince who the fack is dat?

You can calculate 20 years of whatever you like. I've got 20+ years of trading experience minute by minute watching, staring, trading .... the markets.
 
NazSherpa said:
So we can say that a correlation of .63 is somewhere between moderate and strong, fair?
More...
That's fair, but brings us back to my original concern.

Your calculation is looking at the correlation for the same day. Another way of asking my question:

At the close, I see volume is up. So, that means it's likely that today's range was wide. I check, yup, it was. But I could have known that by just looking at the range in the first place.

So how would a correlation, on daily data, between variables on the same day, indicate something about the future?

I'm going to rerun the correlation on spy. But instead of checking it between today's range and volume, I'm going to check it between today's volume and TOMORROW'S range. Be right back.
 
Overnight said:
*spits beverage at screen.* That's 2001! TWENTY YEARS ago! Dude, run it again for the last couple of years?
More...
I ran it for the entire data set that I had available. I'll go with the larger data set, thank you. I'm illustrating that there is a correlation between volume and range. It's one variable that goes into determining the likelihood of a given type of day. If you don't find it useful - don't use it.

A correlation of even .40 in the world of trading is worth paying attention to, IMO
 
SunTrader said:
Vince who the fack is dat?

You can calculate 20 years of whatever you like. I've got 20+ years of trading experience minute by minute watching, staring, trading .... the markets.
More...
I've been trading since '93. Full time since '98, so I guess I've got a year or two on you.
 
You must log in or register to reply here.
Back
Top