Trading Gaps and Breakouts (Stocks)

[vladiator]

Quote from dbphoenix:



Excellent. Since Brandon has so much data collected on the subject, it would be interesting to compare yours with his. Have you broken yours down with regard to size of gap, holding period, minimum necessary stop, etc.?

--Db

Yes, however, unlike him (or maybe b/c I have nothing to promote ), I can't share as many details for obvious reasons. I have filters on capitalization, size of gap that didn't fill (has to be very large, otherwise the effect is not there), abnormal volume, trade size, information flow (e.g. the number of news releases), the timing of those releases etc etc.
The holding period matters to, there are optimal times of entry and exit. The entry time is usually before the close of the drop day, the exit depends on the size of the gap. Could be in extended hours, could be at the open next day, could be later.
Hope this helps.

 

[dbphoenix]

Quote from vladiator:



Hope this helps.

Not really, but thanks for the reply, despite those obvious reasons.

--Db

 

[vladiator]

Quote from dbphoenix:



Not really, but thanks for the reply, despite those obvious reasons.

--Db

Did you want me to give you the exact parameters of the model? Like the gap has to be this % big, the volume this % bigger than normal, etc etc? I spent years finding these and I'm sorry, but I'm just not that altruistic. Secondly, I'm finding that even with the size I'm trading, and even though the stocks are pretty large cap-wise, given the consentrated buying/selling, the price pressure effects are already there. In other words, like most inefficiencies, there are limits to how much capital you can move and having others do what I'm doing wouldn't be helpful at all
Besides, I did give at least the variables to consider, with enough struggle, one can determine the threshold levels for him/herself.

 

[dbphoenix]

Quote from vladiator:

Did you want me to give you the exact parameters of the model? Like the gap has to be this % big, the volume this % bigger than normal, etc etc?

That's pretty much what Brandon has done. That's what makes it useful.

Secondly, I'm finding that even with the size I'm trading, and even though the stocks are pretty large cap-wise, given the consentrated buying/selling, the price pressure effects are already there. In other words, like most inefficiencies, there are limits to how much capital you can move and having others do what I'm doing wouldn't be helpful at all

Given the number of financial instruments available to trade and the number of people likely to imitate you, I doubt this would be a problem. But you never know.

--Db

 

[Brandonf]

Quote from vladiator:



Did you want me to give you the exact parameters of the model? Like the gap has to be this % big, the volume this % bigger than normal, etc etc? I spent years finding these and I'm sorry, but I'm just not that altruistic. Secondly, I'm finding that even with the size I'm trading, and even though the stocks are pretty large cap-wise, given the consentrated buying/selling, the price pressure effects are already there. In other words, like most inefficiencies, there are limits to how much capital you can move and having others do what I'm doing wouldn't be helpful at all
Besides, I did give at least the variables to consider, with enough struggle, one can determine the threshold levels for him/herself.

I think the specifics are generally something that each person has to work out for themselves unless you are going to do a 100% mechanical approach. There are probably as many ways to be a successful trader as there are successful traders, and I really doubt anyone who just follows someone ever truely does well, they can certainly never call themselves a trader. Even with the chatroom, I would not encourage people to be in there relying on us for trades for more then a few months. At some point you have to tailor it to yourself. So in the time between that the objective is just to learn as much as you can so you can do your own thing. I am a big believer in what Richard Dennis said though as well, that you could print your entire method on the front pages of the NY Times and it would hardly matter, most people are not going to be disciplined enough to follow it.

Brandon

 

[vladiator]

Quote from Brandonf:


I am a big believer in what Richard Dennis said though as well, that you could print your entire method on the front pages of the NY Times and it would hardly matter, most people are not going to be disciplined enough to follow it.

Brandon

You are probably right. that's why it's currently undersubmission to an academic journal (at least a pruned down version of it).

 

[vladiator]

Quote from dbphoenix:


Given the number of financial instruments available to trade and the number of people likely to imitate you, I doubt this would be a problem. But you never know.

--Db

Once you screen for everything (it's pretty mechanical and the tests are quite restricting), you'll be suprised at how few instruments remain. It's not unusual to only have 2 setups in a week.

 

[dbphoenix]

Quote from vladiator:



Once you screen for everything (it's pretty mechanical and the tests are quite restricting), you'll be suprised at how few instruments remain. It's not unusual to only have 2 setups in a week.

Unfortunately, I can't comment since I know nothing about what you're doing. I look forward to the publication of your work, assuming it's accepted.

--Db

 

[CalTrader]

Quote from vladiator:



You are probably right. that's why it's currently undersubmission to an academic journal (at least a pruned down version of it).

Hmm ... The methods did not appear to me to be anything that needed to be published ... But I suppose you could surround them with a study of some sorts .... Some of the computer work might be worthwhile to write about but I think that this stuff is old knowledge these days ......

Which journal, article title, and author ?

 

[vladiator]

Quote from CalTrader:



Hmm ... The methods did not appear to me to be anything that needed to be published ... But I suppose you could surround them with a study of some sorts .... Some of the computer work might be worthwhile to write about but I think that this stuff is old knowledge these days ......

Which journal, article title, and author ?

I'm a PhD student, publications are a GOOD thing to have . The reason why it "needs" to be published is b/c it addresses how the confluence of microstructural and behavioural aspects (e.g. overreaction to private information and underreaction to public info (see Hong and Stein, Journal of Finance 1999 or Daniel, Hirshleifer and Subrahmanyam, JF 1998)) can lead to pockets of inefficiency that appear to be exploitable by mechanical trading rules.
It will be an academic finance journal, which one and what author - for privacy reasons, I'd rather not say at this point.