IV Rank

[Shay]

stepandfetchit - thanks for your answer,
yes, i want to get specific.

as Tastytrade said, IV Rank was a game changer for them,
and i want to see it myself, i want to do backtests, and see if they are right (i don't like the tests they did where the starting year is 2009, because in my opinion we have to check everything on 2008 as well), and also - if i can improve their results (maybe specific IV Rank is better..)
i want to calculate IV Rank for all liquid products, for all the options available,
because i think that each day, there are many opportunities, and i want to grade them,
and than i could enter the ones with the most ROC, with the most theta decay, with the highest probability, ... (i know that some parameters are contrary).

i am going to put everything in the Database, do backtests, write different parameters for each trade including General IV Rank, Specific IV Rank, IV Rank of 12 months, 3 months, theta, gamma, ...... )
and then i am going to check with Neural Network to see if there are parameters which stands above the others, giving me extra hedge on something which is already profitable as we seen on TastyTrade (and then i might have more appropriate tool than hand grenades to keep armadillos...)

i want to do that, because every day, we can enter many products, with many expiration dates, with many strikes, with more than one strategy (Naked, Spreads, Calendars, ...)
and how i want to compare my pot-odds between the opportunities.


you attached an image of specific IV for SPX, and you said it comes from the VIX, that make sense,
but how can i get specific IV for AAPL (or any other product) for example ?

Shay

 

[Shay]

by the way, i asked the research team of TastyTrade,
and they said - "We use the IV of the underlying, not of a specific strike. This IV is what is used for the IV Rank Calculation".

Shay

 

[newwurldmn]

I would ignore most of what tasty trade says.

IV rank isn't really that helpful. iv rank is a backward looking statistic while IV is a forward looking statistic. Typically there is a good reason why the rank is high or low.

 

[Shay]

I would ignore most of what tasty trade says.

IV rank isn't really that helpful. iv rank is a backward looking statistic while IV is a forward looking statistic. Typically there is a good reason why the rank is high or low.

i agree that the calculation is problematic,
but the results they presents over and over again, with or without IVRank, are very different in favor of using IVRank.

 

[newwurldmn]

i agree that the calculation is problematic,
but the results they presents over and over again, with or without IVRank, are very different in favor of using IVRank.

Just note that the concept of IV rank is not new or innovative. Pretty much every volatility strategy report has this metric in it.

 

[stepandfetchit]

Shay:
Some more ramblings from me here.
Regarding question: "but how can i get specific IV for AAPL (or any other product) for example ?" From tos option chain as below for specifics:

or from VXAPL for it's "30 day IV -- same derivation as VIX for SPX"
Some of the bigger names have a 30 day IV listed, but not all.

Below is a mapping of Underlying TO 30 day IV symbol that I have found (probably not exhaustive).

# The IV_Map returns the desired IV symbol related to
# the KEY provided.
%IV_Map = qw (
SPY VIX
SPX VIX
IWM RVX
RUT RVX
USO OVX
GLD GVZ
SLV VXSLV
FXE EVZ
EEM VXEEM
EFA VXEFA
EWZ VXEWZ
FXI VXFXI
GDX VXGDX
XLE VXXLE
AAPL VXAPL
AMZN VXAZN
GS VXGX
GOOG VXGOG
IBM VXIBM
OEX VXO
_____________
Note: I use the VIX for SPX and SPY and make similar compromise for IWM to use RVX which may not be as precise as you want, but was close enough for my interests.

Regarding your work load ahead. Seems to me to be a large and daunting task.
One issue with how TastyTrade uses IV rank is they seem to make the assumption that the IV will tend to revert toward some mid-point of the prior 52 weeks IV "during the time of the trade they plan". For example, if it is low, they consider odds are higher that the IV will increase during there trade period, and visa versa if high. Sometimes this occurs, but is possible you may discover some "meat" to add to their "potatoes"! ;-) If one's trade time frame is short, the emphasis on IV rank should diminish accordingly.

Also, below is the "Options Statistics" info from TOS on AAPL showing IV and IV Rank, which is done similar to the CBOE white paper on VIX.

The derivation of the VIX may be found here:
"https://www.cboe.com/micro/vix/vixwhite.pdf" <-- this is well written and provides precise detail on how the VIX is derived from the option chains.

Keep me posted on your progress/findings.

Regards

 

[Dolemite]

i agree that the calculation is problematic,
but the results they presents over and over again, with or without IVRank, are very different in favor of using IVRank.

IV rank is used in hushed tones of reverence by Tastytrade as if it is the holy grail of option selling. Like newwurldmn said, it is usually high (especially very liquid underlyings) for a reason. Their results/tests they perform are often over too short of a time period to be meaningful.

 

[Shay]

Their results/tests they perform are often over too short of a time period to be meaningful.

this is why i want to test it myself

 

[Shay]

stepandfetchit -
yes, i know about the IVs in TOS,
and i know there are some VIX like for other symbols (but not for all !)
but as i said, i want all the data exported to MSSQL
so i want to calculate the data, so every symbol will have IV Rank and not just those with VIX like.

ivolatility and historicaloptiondata (in Level 3) have calculated IVs that i think will be a good starting place for me (instead of me calculate B&S model myself),
so i think i will continue with my coding now, and when i buy the data from these sources, i will see myself the effect of the IVR.

Thanks,
Shay

 

[stepandfetchit]

Shay:
I like what you are doing! Curious if you have resolved your IV concerns satisfactorily. You mention "instead of me calculate B&S model myself" as though that may be difficult. It is not difficult, and the formulas are available. Since you are "digging deeply", it may interest you to add the B&S results as well, to insure you have all available data. Note: Typically, the "standard" B&S models do not account for Dividends, but the modification to incorporate dividends is not difficult.
Regarding IV rank and IV percentile ... I now see that TastyTrade and TOS differ on the definition of IV Percentile! (TOS uses the terms interchangeably, so is possible TastyTrade has more precise definitions, however, both seem to agree on IV Rank, and that IV Rank is more significant a metric).
I also am very interested in this area, but am approaching it from a slightly different perspective.
Have you looked at the CBOE White Paper on VIX? If not, it may remove the uncertainty on how the "standard" Implied Volatility number is derived.
It may be enlightening to determine how "ivolatility" determines the IV numbers in your level 3 data. (Note: I do NOT have access to ivolatility, so am curious). My guess is they reverse engineer the IV from the price of each option (the "knee jerk" approach to this task). This "should" correlate with TOS's "individual implied volatility" derivation. However, it is also possible they "smooth" the iv data by examining the option pricing for all options at the specific strike to approximate a volatility smile curve, which is the preferred method by some option traders.
Pardon all the questions, but this is very interesting stuff!