Search results

So why don't we go over some of your "obvious facts" and take them apart. Options with higher volatility are, surprisingly, also more volatile. This means that it makes more sense to measure the payoff not in percent, but in number of standard deviations needed to break even the option price...

I am sorry, are you a mathematician or a "senior programmer that has used quantlib"? PS. I am kinda suspecting that you are simply taking a piss and I am falling for it...

I am nearly rolling on the floor. Are you bwolinsky's twin? If by "quant" you mean "quantitative trader" then yes, I am a quant. Oh, and yes, I know what quantlib is :)

Well, that's almost as tight as I used to make them for funds. Not too bad, not too bad.

Are you going to say that you are a quant now? Ouch!

When you say they have good spreads, what do you mean? Because I just looked and they have 1 month simple European digital (1300 strike) as 10-wide - is this because of the weekend? Also, the names of the indices are deliberately changed - do they guarantee that the settlement/intraday prices...

Hmm. I take it nobody would do equity SN or index?

You can always simply offset the position by buying a digital with the same strike but an opposite direction, even if you can't sell this one. E.g. FB was at 18 and you bought a digital that pays if FB in one week is at 20. Let's say a day passes and FB is at 20.5 - even if you can't sell the 20...

Obvious Fact: Some people should not be allowed into a library unattended

If you think EOD data is dirty and hard to handle, you should try dealing with tick-level historical options data. There are terabytes of quotes/trades and you have to create vol surfaces out of it. Fun stuff, yet there is a lot of $$ value in getting it all cleaned up and working properly.

It's a "worthless" because it does not add any value to the optionality of the contract. It simply means that you can liquidate your position for the best current bid or put on an offsetting position at the offer side. Unlike for vanilla options, where American feature is usually worth very...

Right, that was my question - I knew that CFE disrupted for system reasons and my PB did not bother me about my positions. So it had to be driven by the broker - was this MtM dislocation/problem driven by IB alone or by other retail brokers too?

"automaton", if i recall correctly, is a "self-operating machine", not a model of any sort.

You are pricing an option and then calculating a real-life payoff, how could the only variable pricing input be "irrelevant"?

Yup. It's no different then delta-hedging a call spread, you end up short gamma when it's OTM and long gamma when it's ITM.

Am I to understand that, as per the other thread, your simulation uses historical volatilities, not real-life implied volatilities?

Actually, a relatively simple back-test will show you that straddles for high beta, high vol stocks are a better buy then straddles on low-beta low-vol stocks, if traded in beta-neutral ratio. In general, the lower the vol, the higher is the proportional risk premium.

Well, they are pricing a digital by pricing a tight call spread, but don't always hedge it. When I was running an exotics book, I would only hedge really large digital risk (e.g. 100m in size) and smaller stuff I would just throw into the book (it is actually booked as a call spread) and...

What do you mean by "more realistic"? Implied volatility is frequently very different then realized volatility, it's part of what option trading is all about.

It's much simpler. A binary option is perfectly replicated via a call spread, that's how a dealer hedges. The bid/offer on the digital is determined by the width of the call spread and the direction of the barrier shift.