Accuracy of IV for Expected Move (for CALLS only)

[raf_bcn]

Hi

The critical thing to remember about IV is that it represents the cost of manufacturing an option via dynamic delta-hedging. It does not represent the expected move of the underlying.

please, is that correct ?

The market maker does not care if the stock moves up more than apparently implied by IV of a call, only that the cost of delta-hedging that move is less than the price he/she sold the option for, on average.

What is the cost of delta hedging? interest, comissions, dividends. The cost of delta hedge changes along one hour? iv does.

 

[stepandfetchit]

just curious,
can you use / recall ATM IV instead of VIX from TOS?
either EOD or intraday..

ThinkScript support for options is horrible at best. You need to be able to programmatically create the OPRA code, which alone will normally exceed the complexity limit of ThinkScript. There is some support for using ATM references which I have done in the past for auto-backtesting Straddles in ThinkScript, but is more problematic than fruitfull. However, for what the OP seems to be desiring, I don't think it is required, at least to the point of determining if there is a "Coon up the Tree" he is currently barking up. -- Using intraday is limited in history, so losses value except for trivial timeframes IMO.

 

[Aquarians]

>> ThinkScript support for options is horrible at best.

Everyone's support for options is horrible at best. This is just much the best moment to challange Black-Scholes of the 70s as it was their turn to challange Binomial of the 50s or the pre-war (as in generalized I'll kill you all if you try me) Bachelier.

 

[Aquarians]

And perhaps coincidentally, my name spells like Bueono.

 

[stepandfetchit]

I coded a simple ThinkScript to perform the following on SPX.
Using VIX as the IV reference, determine the 1-STD up-side price target for 30 Calendar days into the future. For each day, step until that target time and set flag if HIGH SPX price exceeds that value. Repeat for all BARs in Time frame (5Yr picked for this run). Sum number of exceptions, and report as "Price Exceeded up-side IV estimate". This run finds 69 counts. Note, each day is considered a Cannon-Launch trade for the 30 Calendar day period, first failure for each 30 day window counts as one occurrence.
Graph below with the running up-side limit for a rough visual of the threshold.
View attachment 185878

 

[stepandfetchit]

Updated pic identifying each event.

 

[stepandfetchit]

Regarding the heading for this Thread "Accuracy of IV for Expected Move (for CALLS only)"
The "actual" accuracy is dismal! However that does NOT imply no value in using IV for expected move. (It is a forward looking metric) The snip I produced infers that out of 1239 Trades (30 calendar day intervals), only 69, or 5.6% exceeded the 1-STD Dev target on the high side! This is significantly off from "statistical expectation".

 

[Aquarians]

>> I coded a simple ThinkScript to perform the following on SPX.

If anyone here is to have a chance against the Wall Street boyz, you're gonna have to do better.

Complex at least because simple doesn't cut it.

 

[Aquarians]

See my "leap of faith" thread in "programmers" section where I'm offering $1000+ / month for a me++.

(Which obviously means I'm ripping my enhanced version off

 

[Aquarians]

It's not charity and you get 40% equity, therefore I'll enjoy ripping you off, exactly what the world so far has done to me.

I'll post more rigorous differentiation criteria in the next days.