Historical probablility based exits ?

A

Aston01

With access to historical data we have the benefit of being able to go back and look at what happened in the past and compare it to similar events see how common an occurrence this was. I have seen a lot of different exit strategies, but most revolve around some concept of stops, percentages, indicators and etc.

I was wondering if anyone uses or has heard of a method that involves probabilities ?

My thinking on the subject is with information like this you could determine that yea based on the size of this particular move I should probably tighten up my exit strategy a bit because the trade has entered an uncommon area.

I try not to reinvent the wheel if possible so I was hoping to hear if this was an established concept with I name I could further research.
 
This is how a LOT of hedge funds and other quantitative algorithmic trading works. Your not going to read about ways to make money though..if someone is really making money they are not going to share it because the opportunity would be gone in a second.
 
Quote from Aston01:

With access to historical data we have the benefit of being able to go back and look at what happened in the past and compare it to similar events see how common an occurrence this was. I have seen a lot of different exit strategies, but most revolve around some concept of stops, percentages, indicators and etc.

I was wondering if anyone uses or has heard of a method that involves probabilities ?

My thinking on the subject is with information like this you could determine that yea based on the size of this particular move I should probably tighten up my exit strategy a bit because the trade has entered an uncommon area.

I try not to reinvent the wheel if possible so I was hoping to hear if this was an established concept with I name I could further research.
More...

So it's like saying: "I bought this stock and it already went up 10% in 3 hours (an unusual price move historically speaking) so I should exit or lighten up my long position", right?

I don't think it is a good idea, putting an artificial limit on your profits does not pay in the long run.
 
Quote from tradingjournals:

I did work in this area. Do you think there would be interest in it?
More...

I would be interested in knowing where I could possibly find more information on the subject or even a search to term that might narrow down my search.



Quote from xelite777:

So it's like saying: "I bought this stock and it already went up 10% in 3 hours (an unusual price move historically speaking) so I should exit or lighten up my long position", right?

I don't think it is a good idea, putting an artificial limit on your profits does not pay in the long run.
More...

I am not necessarily trying to convert an open ended strategy into a profit target, but my thinking revolved more around the idea that if you took historical precedent into account you might be more aware of the value of the bird in hand relative to the potential of the two in the bush.

... basically more "Money Ball" and less Home Run Derby
 
Quote from Aston01:

I would be interested in knowing where I could possibly find more information on the subject or even a search to term that might narrow down my search.





I am not necessarily trying to convert an open ended strategy into a profit target, but my thinking revolved more around the idea that if you took historical precedent into account you might be more aware of the value of the bird in hand relative to the potential of the two in the bush.

... basically more "Money Ball" and less Home Run Derby
More...

Could you give us a specific example, your question is still rather vague?
 
If you went back and looked at a performance report on a series of trades your going to see things like ... Run-up, drawdown and various efficiency figures to describe how effective the your logic was in capturing the profit the trade was exposed to or limiting the risk.

This is all highly informative, but at the same time a lot of exit methods are adjusted (optimized) to improve upon these metrics yet lack any awareness of historical precedent. So you end up with a "Dumb" system/methodology the does what it is told to do down to the letter of the law without any situational awareness.

What I am looking for is a way to put things into context to where it looks back and says (I'm speaking in generalities here)

...Currently your up 25% which based on similar situations in the past means that there is 5% chance that the trade goes up from here. Knowing this I can decide either "Let it run" or "My odds are too low lets tighten up on this trade an make sure we capture the bird in hand"
 
This is an established method. You will need to do some statistical analysis. See an example in this blog.

Another way is to use a measure of probability like they do in this blog but the details are not clear. In the case of long positions for example you could slowly unload as the probability measure decreases and for short positions you cover slowly as it increases.

Regardless, your idea is sound and this is what most pros do.
 
For me, developing a consistently profitable trading plan was based purely on probabilities of price movements within price action environments that, historically, have repeated again and again (strong trend, channeling trend, wide range, narrow range). The fact is: 95% of aspiring traders don't do this research in a thorough and objective manner.
 
Quote from ronblack:

This is an established method. You will need to do some statistical analysis. See an example in this blog.

Another way is to use a measure of probability like they do in this blog but the details are not clear. In the case of long positions for example you could slowly unload as the probability measure decreases and for short positions you cover slowly as it increases.

Regardless, your idea is sound and this is what most pros do.
More...

Thanks Ronblack, that was definitely in the ball park of what I am looking for. Do you happen to know if this type of analysis has a conventional name? I would sure like to be able to find more information on the topic.

Quote from NoDoji:

For me, developing a consistently profitable trading plan was based purely on probabilities of price movements within price action environments that, historically, have repeated again and again (strong trend, channeling trend, wide range, narrow range). The fact is: 95% of aspiring traders don't do this research in a thorough and objective manner.
More...

I very much agree with you on this point, but in fairness if you look at the majority of general conversations on trading, although they involve math, I wouldn't say they revolve as much around basing decisions on the principle of mathematically verifiable decision making. Don't get me wrong, there is definitely a faction that does take this into account, but it seems like the group making event based decisions is the larger majority.
 
You must log in or register to reply here.
Back
Top