Forums  > Trading  > Expensive put options  
     
Page 1 of 4Goto to page: [1], 2, 3, 4 Next
Display using:  

Scotty


Total Posts: 726
Joined: Jun 2004
 
Posted: 2004-09-14 03:32

This paper suggests that ATM/OTM put options are over-priced by 40-95%.

Put Options

1.  Any thoughts on this issue?

2. Is it possible to replicate and capture this mis-pricing?


“Whatever you do, or dream you can, begin it. Boldness has genius and power and magic in it.”

filthy


Total Posts: 1264
Joined: Jun 2004
 
Posted: 2004-09-14 14:59

not to sound like mikebell but while i haven't read the paper i have an opinion already.

i think puts (otm) are overpriced in practically all cases with equity indices. i have never found a distribution or process that fits the underlying well and gives a skew anywhere near as large as is seen in the options. i think the skew is largely there for microstructure reasons.

but with regard to your second question, why do you need to replicate anything? why not just sell the puts?


"Game's the same, just got more fierce"

AndyM


Total Posts: 2333
Joined: Mar 2004
 
Posted: 2004-09-14 15:37
So this wizard concludes that selling puts between August '87 (S+P: 320 Wink and Dec '00 (S+P: 1320) [with a pretty orderly upward path] would have generated extraordinary profits 'incompatible with the canonical asset-pricing models'. I'm speechless.

My karma will run over your dogma.

tripitaka


Total Posts: 982
Joined: Jun 2004
 
Posted: 2004-09-14 16:53
the question is, how did his model make out selling ATM Nasdaq puts in 2000?

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2004-09-14 17:26

LOL... I just looked at this... ROFL Applause We should have a humor section in the library and put that thing there Smiley


The Figs Protocol.

Scotty


Total Posts: 726
Joined: Jun 2004
 
Posted: 2004-09-14 22:05

1. You could sell puts but then I'm looking for riskless arb.

2. FDAX - why? 


“Whatever you do, or dream you can, begin it. Boldness has genius and power and magic in it.”

Strange


Total Posts: 1597
Joined: Jun 2004
 
Posted: 2004-09-15 01:33

I agree with FDAX - pretty funny stuff. I happend to know a guy that is making money by selling gutted double diagonals on a similar premise: "You can not predict the future, but one thing is certain - theta".


"In Russia, every CDS ends in bullet payment"

Scotty


Total Posts: 726
Joined: Jun 2004
 
Posted: 2004-09-15 02:51

Mkay...So the index increased in value at around 8.5% over the fourteen years. 

I don't think that necessarily invalidates the possibility that a strategy of selling puts is profitable.


“Whatever you do, or dream you can, begin it. Boldness has genius and power and magic in it.”

opmtrader
Founding Member

Total Posts: 1333
Joined: Mar 2004
 
Posted: 2004-09-15 06:44
What is a gutted double diagonal?

or is it like shopping at Tiffany's ... if you have to ask you can't afford it!

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2004-09-15 09:03

opmtrader,

A double diagonal is another term for strangle swap. Say, you buy the front month strangle and sell a back month strangle. A gutted strategy is where you trade the call(s) on the lower strikes and the put(s) on the higher strike, thereby locking in the difference between the two strikes (X2-X1). This part of the strategy is essentially a zero bond, as no matter where S ends up, you will always receive at least (X2-X1).

Scotty,

As Andym already said, the index went from 329.80 to 1320.28  over that period, pretty much in a straight line (with accelerating steepness). You need to write a 42 page paper explaining that you come to the conclusion that puts were overvalued, from a historical perspective?


The Figs Protocol.

Johnny
Founding Member

Total Posts: 4333
Joined: May 2004
 
Posted: 2004-09-15 09:19

Yesterday I read the abstract, but not the paper. The author seemed to be trying to make the point that no reasonable degree of risk aversion could explain the prices of the puts. But mostly, they were trying to use the paper to introduce a new pricing methodology. Maybe when I've cleared through this enormous stack of work I'll get around to reading the paper. Or maybe not.


Stab Art - Capital Structure Demolition, LLC Radiation

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2004-09-15 09:27

He's refering to the "Peso Problem", whereas a catastrophic low probability event could occur but is not in the sample (because of it's low probability).


The Figs Protocol.

Johnny
Founding Member

Total Posts: 4333
Joined: May 2004
 
Posted: 2004-09-15 09:30
RIght, but he says that even taking into account the Peso problem, then no reasonable set of investor preferences could account for the put valuations. This looks interesting to me. But then, as I said, I haven't read the paper so his methodology might be bs ...

Stab Art - Capital Structure Demolition, LLC Radiation

AndyM


Total Posts: 2333
Joined: Mar 2004
 
Posted: 2004-09-15 09:50

Well, which set of investor preferences could account for the valuation of the underlying? By stuffing an extremely unrepresentative set of mkt data into an equilibrium framework, the author is essentially assuming his conclusions.

Scotty, your arithmetic is a bit out: I make it 11% over 13.5 years (plus dividends); more important is the near-monotonic nature of the rally. Of course, this doesn't invalidate the proposition, but it does nothing to validate it either.

(I admit, I haven't read the whole paper; maybe he has something interesting to say, but part 1, which is the cornerstone of the paper, struck me as extremely silly


My karma will run over your dogma.

Johnny
Founding Member

Total Posts: 4333
Joined: May 2004
 
Posted: 2004-09-15 10:27

Well, which set of investor preferences could account for the valuation of the underlying? By stuffing an extremely unrepresentative set of mkt data into an equilibrium framework, the author is essentially assuming his conclusions.

It's a case of looking for a GE solution to the simultaneous valuation of the underlying and the puts, not just of the underlying. This is a perfectly respectible approach, taken by Alan Lewis in his Stoch Vol book, amongst others. To be honest, I would rather see more GE and less RN in general. fuk, now I'm gonna have to read the paper ...


Stab Art - Capital Structure Demolition, LLC Radiation

baghead


Total Posts: 868
Joined: Sep 2004
 
Posted: 2004-09-15 10:54

probably, the most suprising paragraph in finance... ever

page 10

Buying put options is a bearish strategy.That is,put returns are low when the market performs well and vice versa.Over the sample period,the level of the S&P 500 Index has risen more than 4 times,from v 0 314.59 to v T *=1312.15.Is it possible,therefore,that selling puts was so profitable simply because of the unprecedented bull market of the late nineties?In other words,maybe selling puts would not work in downward trending markets?

After that the author introduces a negative drift.

excellent idea!! But selling naked downside puts without taking a kind of pathdependency of vol into account??

could FDAX please hide this thread and invite the guy to this forum?

pleeeease!!!!!


"If you're smart enough to see that markets are not efficient, good for you, otherwise, it's better to assume they are." NeroTulip 2004-08-30 22:21 www.nuclearphynance.com

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2004-09-15 11:03

Excellent idea... we should work out a way to get him to trade with us...Big Smile OTC of courseWink


The Figs Protocol.

Johnny
Founding Member

Total Posts: 4333
Joined: May 2004
 
Posted: 2004-09-15 11:10

Hang on a second! There's lunacy afoot ...

The passage that Baghead quotes is perfectly reasonable. The author says that the S&P rose strongly during the time period under consideration, so it's not a surprise that the put selling strategy worked out. So then the author goes on to consider what happens in the case of a decling market. It's all completely reasonable, and goes in the direction of addressing the concerns raised by AndyM.

And then the lunacy occurs ... Baghead suddenly introduces a completely separate idea and criticizes the author for not mentioning it. It's almost as if Baghead were to go to a restaurant, ask the waiter what food is on the menu and then criticize the waiter for not telling him the scores from that day's football matches! Bizarre!

EDIT: It gets worse. I've skimmed through the paper and I notice that the waiter does in fact mention the football scores in section 4.3. The author writes that the results in the current paper are Partial Equilibrium results, and therefore exclude path-dependent GE type results. However, he goes on to say that he is currently making a GE study and that "preliminary results indicate that the projected pricing kernel ... is not sufficiently path-dependent to rationalize historic put returns".

 

 


Stab Art - Capital Structure Demolition, LLC Radiation

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2004-09-15 11:28

Hang on a second, here: The author looks at historical returns of the S&P 500 from 08/1987 to 12/2000. I appreciate that he may be using the proper tools to look at a problem, but in any case, the entire exercise doesn't make much sense to me.

Take section 2.3.3 for example:

To explore this possibility, we perform another exercise. We introduce a negative drift forthe S&P 500 Index and compute the value of the drift that would reconcile the historical put returns. [.....] The results are reported in Table 5, which reveals that a negative drift of -1.5% per month is necessary for the ATM put to break even.

Now, that's what we've been saying all along and then all of a sudden your puts don't look so ridiculously overprice (granted, still expensive, but nearly as much as the author claims)

Something else, taken from section 2.3.3:

Overall, Table 5 implies that one would need to introduce a highly implausible drift to justify historical put returns.

Hmm... okay, but I assume doing the opposite (looking at a timeseries with a drift and assuming that it's entirely plausible) is OK?

Either way, I'm quite happy taking those puts from him at discounts from 60% to 95%..... hell, I'll bid them at 40% discount even, that should get him really excited.


The Figs Protocol.

Johnny
Founding Member

Total Posts: 4333
Joined: May 2004
 
Posted: 2004-09-15 11:35

FDAX, where's the section where he offers to sell you puts at a discount from 60% to 95%?

This is bizarre. He explicitly discusses the problem raised by AndyM (the bull market during the data sample). He also explicitly discusses the issue raised by Baghead (path-dependency). I don't understand why this paper is getting so many brickbats. After all, the strategy of selling puts on the S&P is reasonably profitable. Not the best in the world, but not the worst either. Consider the returns from the Ansbacher fund (a small fund that writes options on the S&P500), for example:

My suspicion is that some of the people criticizing this paper ( a) haven't read it or ( b) haven't understood it.

 


Stab Art - Capital Structure Demolition, LLC Radiation

AndyM


Total Posts: 2333
Joined: Mar 2004
 
Posted: 2004-09-15 11:45

Agree; the key issue is not the excessive drift, but the quasi-monotonic nature of the rally (the triumph of the 'buy on dips' mentality) that has led to these returns.

No-one would investigate the risk premium puzzle with 13 years of data. The put premium puzzle can in some sense be seen as a leveraged version of the former. Thus I believe that you should use, if anything, more data than when investigating the EPP, to eliminate the distorting effects of leverage.

Having chosen (1) a very small dataset, (ii) a spectacularly ill-chosen one at that, no amount of refinements are going to rescue the structure, since the foundations are rotten.

My 0.02 Dec S+P 1000 puts


My karma will run over your dogma.

Johnny
Founding Member

Total Posts: 4333
Joined: May 2004
 
Posted: 2004-09-15 11:51

"Quasi-monotonic nature of the rally"? I think your memory must be playing tricks on you, old friend. In section 2.3.2 he mentions the 87 crash, the 90 crash when Saddam invaded Kuwait, Spring of 94 when the Fed unexpected raised rates and Autumn 98, the Russia crisis/LTCM fiasco. Surely 4 crises in a 13 year dataset does not constitute a "monotonic rally". But anyway, the point is that he explicitly discusses all these concerns. I don't see what the problem is.

What I did think was pretty neat was the way he sidestepped the joint hypothesis problem. Anyone else like this?


Stab Art - Capital Structure Demolition, LLC Radiation

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2004-09-15 11:51

I'm not debating that selling puts continously makes money on stocks. To me this is quite fundamental. (As a devil's advocate: you can look at the same sample (8/87-12/00) in "New Economy" companies, who were massively undervalued and research (and trading) suggest that most of them were undervalued by several orders of magnitue).

He says that puts range from -39% excess return for ATM puts to -95% excess return for DOTM puts.

Anyway, I also don't understand this, maybe you can help me:

For ATM puts to break even (i.e., to have the average excess return of zero), crashes of the magnitude experienced in October 1987 would have to occur 1.3 times per year.

okay, so now we have an upward trend, that we are introducing downward jumps into. Somehow, that doesn't make alot of sense to me. Because he's still saying that looking at this ridiculous uptrend, is really okay, and puts are just too expensive. Just somehow can't get my head around it, sorry Confused Maybe I'm too much of a risk-neutral guy.


The Figs Protocol.

Johnny
Founding Member

Total Posts: 4333
Joined: May 2004
 
Posted: 2004-09-15 11:55

He's saying two things:

First, he's saying that selling puts earned a greater return than can be explained by any of a class of commonly used equilibrium models, such as CAPM. This is his specific practical point and it's hard to disagree with. Selling puts makes money, as you agreed.

Second, he's using the "selling puts" strategy as an example for the neat trick he developed for side stepping the joint hypothesis problem. This is his general theoretical point and I personally think it's pretty neat.

EDIT: There was one thing I objected to slightly, which is that he seemed to be taking credit for (re)discovering the Fundamental Theorem of Finance. But hey, you can't have everything.

 

 


Stab Art - Capital Structure Demolition, LLC Radiation

FDAXHunter
Founding Member

Total Posts: 8372
Joined: Mar 2004
 
Posted: 2004-09-15 12:29

okay, so the paper boils down to "Puts are too expensive, given the historical context." Agreed?

EDIT: Or to be more specific: "Puts seem to be too expensive, given the historical context and several modeling attempts".


The Figs Protocol.
Previous Thread :: Next Thread 
Page 1 of 4Goto to page: [1], 2, 3, 4 Next