Is this true?? Calling option traders!!

[Omrudra]

Hi all,

I came across this article in the forbes website.
Whither Black-Scholes? - Forbes.com

What do you guys think? I read the thread about whether the black-scholes model is outdated but this article poses another view!!

I would dearly like to know how real option traders feel abt this.
TLDR version:

The author a derivatives consultant from Madrid writes abt how two personalities in the trading world , Taleb and Haug criticises the model

"It's not used (even by those who think that they use it).
--It wasn't needed.

--It wasn't original in the first place. " (from the article from Forbes, 2008 by

Pablo Triana Portela)

I am a begineer in the world of options and trading and am a little confused by it.

Please tell me what u guys think!!

Peace

Omrudra:smart:

 

[Rhody Trader]

As a basis for understanding the basics of options pricing, B-S is useful. Beyond that, the assumptions made are problematic.

 

[Omrudra]

As a basis for understanding the basics of options pricing, B-S is useful. Beyond that, the assumptions made are problematic.

Hi,

Thanks for the reply.

So what do you use as a basis of option pricing. Is binomial trees or any of the other models relevant?

 

[grantx]

Omrudra,

Stick with B-S and its variations. Whatever the problems, no one has produced anything better, or have simply replaced the original shortcomings with others. Most use B-S as starting point or reference at a minimum.

Taleb is a maverick and is a bit too far up his own <<snip>>.

I've just read this article and I can say with absolute confidence this is the biggest load of <<>> I've ever read. Omrudra, ignore this article

Grant.

 

[MrGecko]

Omrudra,

If you are looking only to trade options (as opposed to pricing exoctics etc...) then you needn't be able to derive the BS equation or have a full understanding of all the little nuances between the many models out there.

Going through BS is a fruitful exercise however. As a framework for trading simple puts and calls, it should give you a solid understanding of how each input variable affects price, the greeks, and how these all fit together.

Paul Wilmott and John Hull have both written authoritative, but not inaccessible, books that will cover all you need to know and more. If you are really interested (and calculus savvy), I think i have a copies of the original BS and Merton papers from 76 (?) on an old laptop that I can e-mail you.

 

[gooseman]

You'll find pretty much everybody uses B-S (with tweaks) in the markets so if you use another method to price up you're more than likely going to get steamrollered by the volume in the rest of the market.

Yes we all know about the assumptions but if you're a decent trader you can cover some of the risks posed by the assumptions......for blind reliance on formulae and the problems it poses simply get a copy of 'When Genius Failed-The Collapse of LTCM'. Very much highlights what can go wrong....

 

[grantx]

Omruda,

Expanding on Mr Gecko/Gooseman, learning only the basics of B-S (and the greeks) will not necessarily make you a better trader but you will have a better understaning of where your risk lies.

Grant.

 

[Rhody Trader]

Expanding on Mr Gecko/Gooseman, learning only the basics of B-S (and the greeks) will not necessarily make you a better trader but you will have a better understaning of where your risk lies.

Hmmm. Isn't that what I said?

 

[grantx]

John,

I also expanded on yours slightly.

Grant.

 

[TWI]

I am a begineer in the world of options and trading and am a little confused by it

Keep it simple. Do not assume these lunatics are making anything from their rediculousness. Look at the market itself and understand how the market is pricing optionality. Anything else is likely to get you in to more financial trouble than you can survive.

 

[ft338]

the article if it can be called that is a joke. taleb and people like him take pleasure in ridiculing people's work without really coming up with solutions. his basic mantra is to go long vol at every opportunity and wait for a 'black swan' event which would obviously make outsized returns years like this but bleed carry in normal years.

black scholes isnt that complicated when you think about it, especially if you understand prob and stats. its basically the expected present value of future cashflows using the normal distribution. what black and scholes essentially did was to provide a closed form formula and this is not easy given the mathematical complexities involved in integrating the prob density function of normal distribution. obviously its assumptions dont always hold up but as others have said its very useful for understanding the basics in option pricing and how the greeks work.

also if you use bloomberg or other mediums to follow option prices you will be surprised to see how close the market prices (driven by supply/demand and volatility) at times can be to the BSM prices.

 

[Rhody Trader]

black scholes isnt that complicated when you think about it, especially if you understand prob and stats. its basically the expected present value of future cashflows using the normal distribution.

That's a massive part of the problem right there. The distribution of returns is not normal.

also if you use bloomberg or other mediums to follow option prices you will be surprised to see how close the market prices (driven by supply/demand and volatility) at times can be to the BSM prices.

While it's no doubt true that quote options prices (accounting for the spread, of course) are generally close to where BSM would put them, that's not the problem. The problem is that because BSM is based on poor assumptions it tends to understate the volatility portion of the pricing. As such, if you are a writer of options you will not be receiving enough premium for the risk you are taking.

 

[grantx]

John,

"you will not be receiving enough premium for the risk." In this context, how is the degree of risk underpriced or quantified?

Seperately, if one is using the same model everyday, eg B-S or H-J-M, then questions of absolute values and accuracy are not strictly relevant - isn't it the relative values (compared to historical values) which to determine 'cheap' or 'expensive'?

Grant.

 

[ft338]

im not saying that BSM is a great model for valuing options, im saying that it can be useful to understand option theory and even used as a proxy for pricing options. my point is that its easy to ridicule other people's work without coming up with effective solutions yourself.

as for writer of options not receiving sufficient premium for the risk they take on it would be interesting to see the statistics for the proportion of options that actually end up being profitable for the buyer. it may well be true that the implied vol is understated due to the use of BSM but im sure that this understatement can be analysed and corrected.

as for alternatives and i dont know much about this but do market-makers not use other models and methods for pricing options such as monte carlo simulation?

 

[MrGecko]

I think Monte Carlo would just take too long for a MM - and you'd have to do again for each and every greek.

Also, it doesn't work so well on American contracts.

 

[grantx]

ft338,

"do market-makers not use other models and methods for pricing options such as monte carlo simulation."

I think mm's construct a volatility surface and use this as a basis for their quotes. Now, let's assume this (or any other model used my mm's) is more accurate than B-S, ie premiums will be higher. To repeat the point made above, I think you can still use B-S using the mm's (higher)implieds because the implied may be the only significant aspect which is questionable in B-S.

John Hull's invaluable book, Options, Futures and other Derivative Securities (6th edition) contains treatment of the use of various models.

Grant.

 

[Rhody Trader]

"you will not be receiving enough premium for the risk." In this context, how is the degree of risk underpriced or quantified?

Because the Normal Distribution does not properly account for the probability of larger price moves, option pricing under models using the normal curve includes a volatility component which is too low. That means prices are too low.

Seperately, if one is using the same model everyday, eg B-S or H-J-M, then questions of absolute values and accuracy are not strictly relevant - isn't it the relative values (compared to historical values) which to determine 'cheap' or 'expensive'?

You could make that argument if you are spread trading options against each other, but not if you are in outright positions.

 

[Rhody Trader]

as for writer of options not receiving sufficient premium for the risk they take on it would be interesting to see the statistics for the proportion of options that actually end up being profitable for the buyer. it may well be true that the implied vol is understated due to the use of BSM but im sure that this understatement can be analysed and corrected.

It's not a question of what % of options end up in the money or out of the money that is the problem with the understated volatility in the BSM model - specifically in the use of the normal distribution - but rather how much of a loss is made by the writer when those supposedly improbable events happen.

Imagine how big a hit a put writer on Bear Stearns would have taken and you'll start to see what I'm talking about.

as for alternatives and i dont know much about this but do market-makers not use other models and methods for pricing options such as monte carlo simulation?

I'm no expert on these things, but I understand that binomial models (or variants) are popular.

 

[gooseman]

Imagine how big a hit a put writer on Bear Stearns would have taken and you'll start to see what I'm talking about.
QUOTE]

Don't have to imagine Rhody. One poor US MM wrote a shed load of 55 puts in BS the week before 'it' happened.

 

[grantx]

John,

The greater the number of partitions (or intervals) on a binomial model, the greater the correspondence with B-S. Increase the intervals sufficiently and the binomial and B-S values converge.

Grant.