Offering auto-trading long-only options system "sys13"

[botpro]

It's very simple and I am repeating it all the time! You claim that your system can produce profits while tested in a simulated environment using 100% efficient, unbiased, and fair pricing algorithms GBM and BSM. That is impossible in theory. You are basically claiming that you can make money by betting on random coin flips.

Sorry, you are talking academic bs.

If you claimed your system produced the same profit while tested with real historical data that may have many flaws, biases, inefficiencies, I could not oppose your claims as strongly.

And do you lack the knowledge that one can build flaws, biases, inefficiencies etc. into the model, so that it behaves like the real market?
FYI: it is unneccessary to do that because GBM has it already in it. It is just you who does not understand how GBM, and BSM as well, do work, what the goal of a system is, etc.

So, I think you can't help, and I'm not interested in your "help" or whatever.
End of this fruitless discussion.

PS: here's a practical small exercise for you to finally grasp this simple stuff:

You buy 1 call when spot is 100:
./BSM.exe 100 100 30 60
Spot=100.00 Strike=100.00 AnnVola%=30.00 ExpDays=60.00(t=0.23810) AnnDrift%=0.00 AnnDivid%=0.00 --> Call=5.83471 Put=5.83471

And after 5 days the underlying has risen 4% to spot 104 and you close the position:
./BSM.exe 104 100 30 55
Spot=104.00 Strike=100.00 AnnVola%=30.00 ExpDays=55.00(t=0.21825) AnnDrift%=0.00 AnnDivid%=0.00 --> Call=7.91974 Put=3.91974

How much profit have you made?

 

[ospkfne]

When you introduced yourself you wrote that you were a math researcher, so I thought you would be able to comprehend the theoretical reasoning behind what I am trying to explain to you. I even tried to build the simple program for you that would demonstrate and confirm my arguments.

Anyway, I will let you believe that you found a philosopher's stone that turns copper into gold. Good luck selling it to the best and brightest traders at elitetrader.com!

 

[botpro]

When you introduced yourself you wrote that you were a math researcher, so I thought you would be able to comprehend the theoretical reasoning behind what I am trying to explain to you. I even tried to build the simple program for you that would demonstrate and confirm my arguments.

Anyway, I will let you believe that you found a philosopher's stone that turns copper into gold. Good luck selling it to the best and brightest traders at elitetrader.com!

Yes, I'm a math researcher as well, but only non-professional as stated.

But what especially some physists like you from the academia think of the markets is simply wrong.

 

[ospkfne]

... one can build flaws, biases, inefficiencies etc. into the model, so that it behaves like the real market?
FYI: it is unneccessary to do that because GBM has it already in it

Oh, really? Can you be specific and tell us which flaws are already in GBM?

You buy 1 call when spot is 100:
./BSM.exe 100 100 30 60
Spot=100.00 Strike=100.00 AnnVola%=30.00 ExpDays=60.00(t=0.23810) AnnDrift%=0.00 AnnDivid%=0.00 --> Call=5.83471 Put=5.83471

And after 5 days the underlying has risen 4% to spot 104 and you close the position:
./BSM.exe 104 100 30 55
Spot=104.00 Strike=100.00 AnnVola%=30.00 ExpDays=55.00(t=0.21825) AnnDrift%=0.00 AnnDivid%=0.00 --> Call=7.91974 Put=3.91974

How much profit have you made?

Yeah, this particular trade is obviously very profitable, you made 35.73% in 4 days. We all understand that options can sometimes make 2x, 5x, or 10x, some OTM 100x. What you need to change in your thinking though is not to look at individual trades where you got lucky, but rather look for overall "expected value" (EV) of your trade setups. See https://en.wikipedia.org/wiki/Expected_value.

Good practical way of how to estimate EV of any trade setup would be either to replay 1000 different historical data patterns having this setup or in your case repeat the random walk 1000 times between the buy and sell to determine the mean of all the profits and losses. And this is what my little program that I wrote for you is trying to do. It's trying to estimate the expected value of any trade setup you may find in your trading system by replaying the GBM random walk 1000x.

 

[botpro]

Oh, really? Can you be specific and tell us which flaws are already in GBM?

Oh boy! There are no flaws in GBM. GBM already models the behaviour of a stock.

Yeah, this particular trade is obviously very profitable, you made 35.73% in 4 days. We all understand that options can sometimes make 2x, 5x, or 10x, some OTM 100x. What you need to change in your thinking though is not to look at individual trades where you got lucky, but rather look for overall "expected value" (EV) of your trade setups. See https://en.wikipedia.org/wiki/Expected_value.

No, it is exactly you who needs to change his thinking. The market as a whole is a zero-sum game,
but you cannot conclude from that that everybody gets the same equal share of the possible outcomes.

I'm glad you finally understand the leverage effect of options.

I would advise you to discuss your market theory in a different thread as it gets more and more off-topic here. Thx.

 

[dartmus]

ospkfne, given perfect knowledge of the constraints on the constraints within pseudo random data generators it should b relatively easy to take advantage of weaknesses in the data generator. IOW imo it's not possible to write a fair RNG unless the constraints are very large numbers near infinity and don't constrain the next value based upon previous values.

Otherwise 'everything' you said is correct. It's not possible to reverse engineer real data without using real data.

 

[garachen]

Oh, really? Can you be specific and tell us which flaws are already in GBM?



Yeah, this particular trade is obviously very profitable, you made 35.73% in 4 days. We all understand that options can sometimes make 2x, 5x, or 10x, some OTM 100x. What you need to change in your thinking though is not to look at individual trades where you got lucky, but rather look for overall "expected value" (EV) of your trade setups. See https://en.wikipedia.org/wiki/Expected_value.

Good practical way of how to estimate EV of any trade setup would be either to replay 1000 different historical data patterns having this setup or in your case repeat the random walk 1000 times between the buy and sell to determine the mean of all the profits and losses. And this is what my little program that I wrote for you is trying to do. It's trying to estimate the expected value of any trade setup you may find in your trading system by replaying the GBM random walk 1000x.

Please, just put him on ignore. Otherwise, your posts temp me too much to take him off ignore to see what new foolish, arrogant drivel he spews. Then, when I do, I immediately regret it as it is even worse than I imagined. I can't figure out if he actually believes what he's saying or if he's just being contrary as a joke.

 

[botpro]

ospkfne, given perfect knowledge of the constraints on the constraints within pseudo random data generators it should b relatively easy to take advantage of weaknesses in the data generator. IOW imo it's not possible to write a fair RNG unless the constraints are very large numbers near infinity and don't constrain the next value based upon previous values.

Otherwise 'everything' you said is correct. It's not possible to reverse engineer real data without using real data.

Thx for your comment, but please discuss the RNG stuff in a different thread as it is off-topic here.

 

[botpro]

Please, just put him on ignore. Otherwise, your posts temp me too much to take him off ignore to see what new foolish, arrogant drivel he spews. Then, when I do, I immediately regret it as it is even worse than I imagined. I can't figure out if he actually believes what he's saying or if he's just being contrary as a joke.

Same for you: this thread is about a biz-offer made, not about discussing personal attidutes. Thx.

 

[ospkfne]

dratsum, thank you for pointing that out, weak RNG could be a possibility. It would be a good question for botpro: is your system exploiting default_random_engine generator?