Search results

Note the date. I am sure that guy has done perfectly well in August :)

Not really. I think if you are an experienced vol trader, you can easily dedicate part of your portfolio to vix (especially if you have experience with volatility derivatives) and keep doing other things at the same time. I can name a whole bunch of reasons to be long VXX vs some other forms...

There is like a million ways of doing it, depending on what your are trying to achieve. Personally, I think using a risk-neutral probability is the least useful, since it does not really provide you with any _relative_ value, but simply converts implied vol to probability.

There is no real difference in roll-down of the first futures to spot vs "roll-over" of the replicated constant maturity futures which is VXX. For an equivalent weighted vega position, you are going to be locking the same loss/gain over a long enough holding period. Some simple analysis of...

Pros both buy and sell convexity/risk premium. Unless you trading your own capital or have captive investors, you can't afford being either always long or always short risk premium. PS. Given how steep the vol surface is, VXX is not a good buy.

Stop loss is only good if you are sure you can get filled. Otherwise, buy and hold with a stop loss would be a super-strategy too.

Swimming a bit in your terminology, so need some help here. In short you are selling the straddle break-even strike and buying the straddle. As a result, you would probably be slightly long vega and short skew (I have not modeled these). Couple questions: Is the idea that disregarding the...

Generically, strategies involving options (or pretty much any other convex payoffs) are very hard to analyze using conventional performance metrics. If you want to convince yourself that your strategy is bullet-proof, you want to run some historical scenarios. Make sure to include as many crisis...

If your option expires in a year, the spot price is 100, your interest rate is 1% and your div yield is 3% then your stock forward is 100 * exp((1% - 3%) * 1 year) = about 98%.

Yup, that is very true. Every new trade is up to God. However, as you spend more time building models, testing and thinking about risks, God helps you more.

Nope, the more the merrier. For example, lets say you are buying S&P futures when 50 day MA crosses 250 day MA. In this case, your "condition" is the cross between two MAs, that is your "signal". Your ex-condition test would be - what is the return of S&P futures on all days, not only when...

Actually, it's a good question if converts are always long convexity. I think the best way is to think of them as riskless bond + call - put where the put comes from the credit risk. Once you think about it this way, you realize that it's a skew position too :P

That problem has been solved long time ago - you can trade down-and-out puts if you have access to OTC markets. Or, if you do not, you can trade skew flies - i trade these a lot. Or you can trade one skew against another. Or you can trade gamma against long-dated skew as a vega reset trade...

Root-time adjusted it's self consistent with respect to the terminal distributions. Of course, the skew at the short end is a very different position then long-end skew (i think i've shared some of my relative skew ideas with you over bbg). Usually root-time adjusted sk10 is actually upward...

You need to simply multiply the Price by exp(interest rate - div rate).

Being short KOs is always better - you can make a back envelope B/E calculation and short/long some spot. Long KOs suck - a couple times I got stuck just a few inches away from the barrier, it felt like a hot poker up my ass..

were you long the KO or short it? loads of fun either way (especially when you have a quarter-yard of KO a few ticks away)...

even as you adjust by root time? e.g. sk10 = (iv(90%) - iv(100%))*sqrt(t) for a single asset it gives you a very consistent picture of the "term structure of skew" PS. using x-delta risk reversal normalized by ATM volatility is the other "cool" way

This is so misguided at some many levels, I don't even know where to start. Do you at least understand that when you are trading futures, you are subject to the same risk-premium capture (positive or negative) as you are when you are trading ETFs?

I'll try that, interesting idea. Would you use random bar length x=R(1,2,3) or simply generate all possible combinations and use random sample of trades as "out-of-sample"?