Pages:
Author

Topic: Options trading thread - page 3. (Read 5770 times)

legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 10:51:14 PM
#81
Yup. ok.

btw..are there any options trading platforms out there at all?

not that i know of.

icbit.se and mpex.co have futures and mpex had options but now they dont have any market makers anymore, so there is that.
sr. member
Activity: 300
Merit: 253
Ok Check!
April 14, 2014, 10:49:29 PM
#80
Yup. ok.

btw..are there any options trading platforms out there at all?
legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 09:33:31 PM
#79
Lol, just looked again and found http://www.option-price.com/documentation.php

standard black scholes modell. Hmm

yup..so you see the problem is about the dividend yield, the normal return distribution and the vols

actually within 14 days, the dividenyield and intrest rate is pretty much irrelevant.

With NORMALLY DISTRIBUTED returns, the implied volatility of this deal is:



how did you fill in the current market price as 29?

Implied volatility suggest that there is a market price. And as this option I sold was the only one I knew of, I just calulated 0.065BTC/1BTC*450USD = 29USD
sr. member
Activity: 300
Merit: 253
Ok Check!
April 14, 2014, 09:31:49 PM
#78
Lol, just looked again and found http://www.option-price.com/documentation.php

standard black scholes modell. Hmm

yup..so you see the problem is about the dividend yield, the normal return distribution and the vols

actually within 14 days, the dividenyield and intrest rate is pretty much irrelevant.

With NORMALLY DISTRIBUTED returns, the implied volatility of this deal is:



how did you fill in the current market price as 29?
legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 05:18:50 PM
#77
Lol, just looked again and found http://www.option-price.com/documentation.php

standard black scholes modell. Hmm

yup..so you see the problem is about the dividend yield, the normal return distribution and the vols

actually within 14 days, the dividenyield and intrest rate is pretty much irrelevant.

With NORMALLY DISTRIBUTED returns, the implied volatility of this deal is:

sr. member
Activity: 300
Merit: 253
Ok Check!
April 14, 2014, 05:12:21 PM
#76
Lol, just looked again and found http://www.option-price.com/documentation.php

standard black scholes modell. Hmm

yup..so you see the problem is about the dividend yield, the normal return distribution and the vols
legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 04:19:09 PM
#75
Lol, just looked again and found http://www.option-price.com/documentation.php

standard black scholes modell. Hmm
legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 04:17:37 PM
#74

I am currently in a finance-science master program. My field is nearly the same.

I did price the option using a floating variance calculation. This lead to a price of 0.03 ~. Don't have the calculations here though.

You can also check the implied volatility of a given price, you will see that the IV for a far out of the money option, with a short runtime, will be EXTREMELY high, like 200-400% a year. Total crazy.

One thing to note is, that bitcoin is not normally distributed at this moment in time. It is a fattail distribution paradise.

That's why I chose to add a 100% premium on the calculated fair price.


Good to know about your program. I am a graduate of a FE program.  Smiley

When you say floating var, I believe you mean using a stochastic vol? Did you use any of the pre built modeling techniques?

The IVs should come around those numbers. That's how volatile bitcoin has been. This also means that the options will technically be pretty expensive.

That in red is the real challenge in pricing a btc option fairly. So your fair price was from your own calculation or using some online calculator? If you used an online one, would you mind sharing the link of what you used?

Yeah stochastic. Im no native english speaker Wink

I was using a own approach that is working quite well for stocks too. Very similar to the event study approach though.

Options don't need to be expensive, there is a fair price, the problem is, that the best guesses will have a very high error term. That's the reason. Wall street has 30 year+ of data they can ovserve. We got about 2-3 years of good data (good is a faaaaar stretch)

Fattails can be priced, but it is very complicated to do it by hand, I should write a script for that....

I was doing it by hand and by a calculator I found online (It looks quick and dirty tbh) http://www.option-price.com/ . The calculator was pricing the options way too low in my eyes (even below the 0.035). So I don't know, there is no explanation on how it works exactely.
legendary
Activity: 2968
Merit: 1198
April 14, 2014, 04:16:21 PM
#73
Bitfinex is providing such a variable too: https://www.bitfinex.com/pages/stats  129%. Nearly the same value that I got. But that is calculated over 7 days so completely useless for a good indicator.

They should at least graph it. Wouldn't necessarily be predictive but it would be more informative.

sr. member
Activity: 300
Merit: 253
Ok Check!
April 14, 2014, 04:10:14 PM
#72
How to accurately price a bitcoin option? Apart from a willing buyer and seller, the risk adjusted premium should be the real "fair"  value.

There is no such thing as an "accurate" price without some sort of model. What the seller says he did in constructing his offer was similar to what I did in evaluating it. You will notice there is a fair component of vague "it doesn't quite fit the model, so I increased the price." Reality is likely more complex than any model, which means all models will be inaccurate (though often still useful).





I completely agree with you. Accurate is more likely a misleading term. I should have said statistically meaningful price.

The fact that the reality of btc pricing is different than theoretical models makes it all the more exciting. Doesn't it?
legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 04:09:58 PM
#71
How to accurately price a bitcoin option? Apart from a willing buyer and seller, the risk adjusted premium should be the real "fair"  value.

There is no such thing as an "accurate" price without some sort of model. What the seller says he did in constructing his offer was similar to what I did in evaluating it. You will notice there is a fair component of vague "it doesn't quite fit the model, so I increased the price." Reality is likely more complex than any model, which means all models will be inaccurate (though often still useful).





yeah, one extremely big unknown variable is the variance. I have calulated it with an excel spreadsheet. The amount was way too low, but it was very fattailed. So very inaccurate as a good guessing variable.

Bitfinex is providing such a variable too: https://www.bitfinex.com/pages/stats  129%. Nearly the same value that I got. But that is calculated over 7 days so completely useless for a good indicator.
sr. member
Activity: 300
Merit: 253
Ok Check!
April 14, 2014, 04:08:27 PM
#70

I am currently in a finance-science master program. My field is nearly the same.

I did price the option using a floating variance calculation. This lead to a price of 0.03 ~. Don't have the calculations here though.

You can also check the implied volatility of a given price, you will see that the IV for a far out of the money option, with a short runtime, will be EXTREMELY high, like 200-400% a year. Total crazy.

One thing to note is, that bitcoin is not normally distributed at this moment in time. It is a fattail distribution paradise.

That's why I chose to add a 100% premium on the calculated fair price.


Good to know about your program. I am a graduate of a FE program.  Smiley

When you say floating var, I believe you mean using a stochastic vol? Did you use any of the pre built modeling techniques?

The IVs should come around those numbers. That's how volatile bitcoin has been. This also means that the options will technically be pretty expensive.

That in red is the real challenge in pricing a btc option fairly. So your fair price was from your own calculation or using some online calculator? If you used an online one, would you mind sharing the link of what you used?
legendary
Activity: 2968
Merit: 1198
April 14, 2014, 04:04:46 PM
#69
How to accurately price a bitcoin option? Apart from a willing buyer and seller, the risk adjusted premium should be the real "fair"  value.

There is no such thing as an "accurate" price without some sort of model. What the seller says he did in constructing his offer was similar to what I did in evaluating it. You will notice there is a fair component of vague "it doesn't quite fit the model, so I increased the price." Reality is likely more complex than any model, which means all models will be inaccurate (though often still useful).



legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 03:59:45 PM
#68
I wonder how are you guys pricing the option? How do you arrive at a particular premium to be paid to the option seller?

Willing buyer + willing seller.  

I wonder how are you guys pricing the option? How do you arrive at a particular premium to be paid to the option seller?

I offered him a price, which he accepted. The price was 0.065 and it was twice what an option pricing programm gave me as a "fair" value.

So this option pricing program is something which utilizes Black Scholes method? I ask this because fundamentally the bitcoin options can't be priced using the traditional B-S Options pricing formula for other equities and asset classes. What this means is that the option writer is statistically taking more risk than what he is perceiving as per the option calculator.

This has been an area of research for me. How to accurately price a bitcoin option? Apart from a willing buyer and seller, the risk adjusted premium should be the real "fair"  value.

I am currently in a finance-science master program. My field is nearly the same.

I did price the option using a floating variance calculation. This lead to a price of 0.03 ~. Don't have the calculations here though.

You can also check the implied volatility of a given price, you will see that the IV for a far out of the money option, with a short runtime, will be EXTREMELY high, like 200-400% a year. Total crazy.

One thing to note is, that bitcoin is not normally distributed at this moment in time. It is a fattail distribution paradise.

That's why I chose to add a 100% premium on the calculated fair price.

sr. member
Activity: 300
Merit: 253
Ok Check!
April 14, 2014, 03:56:12 PM
#67
I wonder how are you guys pricing the option? How do you arrive at a particular premium to be paid to the option seller?

Willing buyer + willing seller.  

I wonder how are you guys pricing the option? How do you arrive at a particular premium to be paid to the option seller?

I offered him a price, which he accepted. The price was 0.065 and it was twice what an option pricing programm gave me as a "fair" value.

So this option pricing program is something which utilizes Black Scholes method? I ask this because fundamentally the bitcoin options can't be priced using the traditional B-S Options pricing formula for other equities and asset classes. What this means is that the option writer is statistically taking more risk than what he is perceiving as per the option calculator. In case he offered double, then the option buyer paid more for the risk he offloaded. In any case it is technically mispriced.

This has been an area of research for me. How to accurately price a bitcoin option? Apart from a willing buyer and seller, the risk adjusted premium should be the real "fair"  value.
legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 02:27:30 PM
#66
I wonder how are you guys pricing the option? How do you arrive at a particular premium to be paid to the option seller?

I offered him a price, which he accepted. The price was 0.065 and it was twice what an option pricing programm gave me as a "fair" value.
legendary
Activity: 2968
Merit: 1198
April 14, 2014, 02:26:44 PM
#65
I wonder how are you guys pricing the option? How do you arrive at a particular premium to be paid to the option seller?

Willing buyer + willing seller. 
sr. member
Activity: 300
Merit: 253
Ok Check!
April 14, 2014, 02:22:52 PM
#64
I wonder how are you guys pricing the option? How do you arrive at a particular premium to be paid to the option seller?
legendary
Activity: 1680
Merit: 1001
CEO Bitpanda.com
April 14, 2014, 02:09:23 PM
#63
1 day and 6 hours. Please for the love of god. Let me lose this one Cheesy

It's trying...


Would suck for you if it reaches higher prices shortly after the end of the bet. Still better than not rising tough, i bet.
legendary
Activity: 2968
Merit: 1198
April 14, 2014, 02:07:30 PM
#62
1 day and 6 hours. Please for the love of god. Let me lose this one Cheesy

It's trying...
Pages:
Jump to: