Hello all!
I recently came across a trading method where the focus is on intraday or overnight price movement.
I have done a lot of general mathematical research on probability, statistics, etc. in order to further develop this system, or at least try to verify its feasibility.
I tried to write a javascript application to make the necessary calculations and see if the method is a "winning" system. However, I am not a professional programmer, and at this point I am so immersed in it that I don't even know if the application is accurate or not anymore.
---
The following is the method. Please note that I didn't find anything similar when searching the board, but I apologize if I missed a thread about this and I am repeating something or someone.
Genearl Ideas:
1) Observe INTRADAY price movement between OPEN and HIGH (bull) or OPEN and LOW (bear).
2) Alternatively, observe OVERNIGHT price movement between CLOSE and HIGH of the next day (bull) or CLOSE and LOW of the next day (bear).
3) Calculate percentages (probabilities) of how likely a certain price movement is by comparing each price movement to all of the others calculated (less than or equal to, greater than or equal to, etc.)
My idea was to use a long time range of data and see how likely various price movements are to occur. (e.g. maybe a stock has moved up at least 1/4 point from its opening price 90% of the time (days) observed). A person could re-calculate the numbers at the end of each day and place orders the next day accordingly.
I really think that a system like this could yield decent results, considering that it is purely mechanical and EOD. I am particularly interested in such a system because I am a professional and I do not have enough time during the day (nor the patience, really!) to be in front of the computer monitoring the markets.
The questions and concerns that are left unresolved are:
1) What amount of stop-loss or risk management should be used with this system to protect from large movements opposite of the trade?
2) How would a person use win/loss or profit/loss ratios to calculate a mathematically sound stop-loss?
3) Should the position be closed out at EOD if the profit target has not been met?
Again, I am not a mathematician, and I was hoping that since many of you have been around for a while, you may have a good idea of how to manage or improve upon such a system.
Ideally, it would be great if a mathematician would post here and enlighten us all as to how the probabilities would work and how to set risk management so that the numbers are heavily in our favor in such a system. It would be great if such a system would yield a long-term advantage (like a casino, which always comes out ahead mathematically, even if by a little or over the long run).
I personally would be happy with very high probability trades (90%+) of very low profit targets (1/10 of a point, ~1% movement, etc.). Of course, a higher profit target is always OK, haha!! But I am not greedy. I am basically looking for ways to make this system turn fairly consistent profits on a daily basis, although those profits may be very low.
Obviously, I will be checking this thread often, so if anybody has any questions or needs any clarification, please post and I will reply as soon as I can.
I hope we can all benefit, or at least learn a little more, from the ideas contributed about this method.
Thank you all in advance for the help and ideas!
I recently came across a trading method where the focus is on intraday or overnight price movement.
I have done a lot of general mathematical research on probability, statistics, etc. in order to further develop this system, or at least try to verify its feasibility.
I tried to write a javascript application to make the necessary calculations and see if the method is a "winning" system. However, I am not a professional programmer, and at this point I am so immersed in it that I don't even know if the application is accurate or not anymore.
---
The following is the method. Please note that I didn't find anything similar when searching the board, but I apologize if I missed a thread about this and I am repeating something or someone.
Genearl Ideas:
1) Observe INTRADAY price movement between OPEN and HIGH (bull) or OPEN and LOW (bear).
2) Alternatively, observe OVERNIGHT price movement between CLOSE and HIGH of the next day (bull) or CLOSE and LOW of the next day (bear).
3) Calculate percentages (probabilities) of how likely a certain price movement is by comparing each price movement to all of the others calculated (less than or equal to, greater than or equal to, etc.)
My idea was to use a long time range of data and see how likely various price movements are to occur. (e.g. maybe a stock has moved up at least 1/4 point from its opening price 90% of the time (days) observed). A person could re-calculate the numbers at the end of each day and place orders the next day accordingly.
I really think that a system like this could yield decent results, considering that it is purely mechanical and EOD. I am particularly interested in such a system because I am a professional and I do not have enough time during the day (nor the patience, really!) to be in front of the computer monitoring the markets.
The questions and concerns that are left unresolved are:
1) What amount of stop-loss or risk management should be used with this system to protect from large movements opposite of the trade?
2) How would a person use win/loss or profit/loss ratios to calculate a mathematically sound stop-loss?
3) Should the position be closed out at EOD if the profit target has not been met?
Again, I am not a mathematician, and I was hoping that since many of you have been around for a while, you may have a good idea of how to manage or improve upon such a system.
Ideally, it would be great if a mathematician would post here and enlighten us all as to how the probabilities would work and how to set risk management so that the numbers are heavily in our favor in such a system. It would be great if such a system would yield a long-term advantage (like a casino, which always comes out ahead mathematically, even if by a little or over the long run).
I personally would be happy with very high probability trades (90%+) of very low profit targets (1/10 of a point, ~1% movement, etc.). Of course, a higher profit target is always OK, haha!! But I am not greedy. I am basically looking for ways to make this system turn fairly consistent profits on a daily basis, although those profits may be very low.
Obviously, I will be checking this thread often, so if anybody has any questions or needs any clarification, please post and I will reply as soon as I can.
I hope we can all benefit, or at least learn a little more, from the ideas contributed about this method.
Thank you all in advance for the help and ideas!
