Author

Topic: Black Swan Events and Multiple Algos (Read 1648 times)

legendary
Activity: 2408
Merit: 1121
July 03, 2012, 02:46:43 AM
#12
Black-Scholes and the assumption of a (normal) gaussian distribution in determining option pricing is one of the major flaws that blew up LTCM. The 'mean' should be thicker than the 'tails', and most price action disagrees. The number of two to three sigma events fall outside of the 'predicted' distribution, which means of course any assumptions based on that model are doomed to tail-risk of large magnitudes.

BTC is probably a more 'sane' market that anything out in the traded exchanges, if only by virtue of not having a crapload of sub-pennying front-running High-Frequency Trading algorithms dominating volume like the ES futures or the major Index stocks.
hero member
Activity: 728
Merit: 500
June 25, 2012, 09:54:40 PM
#11
Sorry, not picking on you, just pointing out the futility in trying to address some of the stuff. The problem is that the more accurate you want the information, the more technical it's gotta get, which is unfortunately the problem. I only consider myself slightly aware of some of these things, but doing research into why so many financial institutions like the Black-Scholes equation is a good start.

Fair warning though, it gets technical very quickly. There may be some Khan Academy tutorial that kind of gives a general overview, though.

http://en.wikipedia.org/wiki/Kurtosis_risk
http://en.wikipedia.org/wiki/Black%E2%80%93Scholes
http://en.wikipedia.org/wiki/Delta_hedging
http://www.youtube.com/watch?v=zFiGL7lxsxc

Haha I was just answering your question.
donator
Activity: 1419
Merit: 1015
June 25, 2012, 09:41:21 PM
#10
Sorry, not picking on you, just pointing out the futility in trying to address some of the stuff. The problem is that the more accurate you want the information, the more technical it's gotta get, which is unfortunately the problem. I only consider myself slightly aware of some of these things, but doing research into why so many financial institutions like the Black-Scholes equation is a good start.

Fair warning though, it gets technical very quickly. There may be some Khan Academy tutorial that kind of gives a general overview, though.

http://en.wikipedia.org/wiki/Kurtosis_risk
http://en.wikipedia.org/wiki/Black%E2%80%93Scholes
http://en.wikipedia.org/wiki/Delta_hedging
http://www.youtube.com/watch?v=zFiGL7lxsxc
hero member
Activity: 728
Merit: 500
June 25, 2012, 09:27:57 PM
#9
I want to learn about Kurtosis risk!

Explain it like I am 5...

Tongue

Its the risk someone will smack you in the back of the head when you least expect it and you'll bite that tongue off.
donator
Activity: 1419
Merit: 1015
June 25, 2012, 09:21:23 PM
#8
I want to learn about Kurtosis risk!

Explain it like I am 5...

Tongue
hero member
Activity: 728
Merit: 500
June 24, 2012, 03:36:35 PM
#7
The data from the trading glitch has actually produced some extra divergence in my models—half a percent or so. But I don't like to throw out data, it will take me a little bit to find a way to work it out.

Thank you for your acknowledgement. I am 12 (years old?) and appreciate it.

It just means I'm new at it.
legendary
Activity: 1022
Merit: 1000
June 24, 2012, 12:47:48 PM
#6
The data from the trading glitch has actually produced some extra divergence in my models—half a percent or so. But I don't like to throw out data, it will take me a little bit to find a way to work it out.

Thank you for your acknowledgement. I am 12 (years old?) and appreciate it.
hero member
Activity: 728
Merit: 500
June 24, 2012, 01:19:54 AM
#5
The data from the trading glitch has actually produced some extra divergence in my models—half a percent or so. But I don't like to throw out data, it will take me a little bit to find a way to work it out.

Thank you for your acknowledgement. I am 12 and appreciate it.
hero member
Activity: 728
Merit: 500
June 24, 2012, 12:32:26 AM
#4
I think they are recalibrating right now...
hero member
Activity: 728
Merit: 500
June 23, 2012, 10:38:31 PM
#3
Do the events of earlier today put you at a disadvantage?

You are referring to the unintentional halt in trading at Gox?
Yes, sorry.
full member
Activity: 157
Merit: 100
June 23, 2012, 10:37:39 PM
#2
Do the events of earlier today put you at a disadvantage?

You are referring to the unintentional halt in trading at Gox?
hero member
Activity: 728
Merit: 500
June 23, 2012, 10:33:09 PM
#1
So there are at least a few people here running multiple models that they pick and choose to trade based on, whether using computers or elliot wave theory. Do the events of earlier today put you at a disadvantage?

It seems to me every time one of these technical difficulties occurs the market becomes "wary" and kind of resets, negating recent info that has been included in the models.

I'm just interested to know how people deal with that situation.
Jump to: