|
Boris Lund
 Regular
  Posts: 59
Joined: 12/9/2021
User Profile |
I am very interested in expanding and combining your two flagships ARM and OptionsFinder.
How many functions can we possibly have from options? PutV and CallV is a very crude resolution of the options market. It would be very useful to be able to pull value and open interest for both puts and calls for set strike prices or even to find out where certain percentiles of value are located for a given expiry date. Further, being able to tell when the next large gamma expiry is happening would enable us to “see” the structure of options positioning for a given symbol and a given period much better.
Options finder does much of this when searching for good prospects but we need to be able to access this information in scripts.
I am inspired by spotgamma who provides spy/spx data together with an analysis, and I would very much like to do some ARM training based on the same data. Spotgamma will tell you the expected volatility based on the gamma together with resistance/support levels based on hedging.
Being able to see strike prices with unusually large hedging or just put or call values with dates tells us a lot about market levels and structures in the near future.
And, in addition, knowing when large amounts of gamma expires would be invaluable.
In an ARM context, coding this information as genes allows us to look for shapes and general technical patterns but with the added advantage of knowing whether or not to expect such patterns to develop freely or if sudden movements due to large hedging or options-rollling movements are to be expected to interfere.
Could you expand on documentation on the possibilities with the existing function calls that come with option finder and are accessible in omni language and supply the last levels of information for scripting which option finder already looks at?
/best regards Boris
|
|
Jim Dean
 Sage
       Posts: 3433
Joined: 3/13/2006
Location: L'ville, GA
User Profile |
Hi Boris. I understand your idea and see it’s value.
One important thing to remember about Options that differ from Stocks - Nirvana uses “fake” historical Options data for option finder - it’s really good, plausible fake - since trying to gather, store and process the HUGE number of actual historical datapoints related to the many different contracts and strike prices of thousands of symbols is just beyond the practical scope of the local PC and data feed. Also, it’s pretty expensive to purchase from data vendors.
Since that data is not “real”, then any ARM processing on it would be compromised significantly, especially for outlier conditions that many multi-leg options trades utilize.
It might be possible for N to do this on a cloud server (or ten or a hundred). I’m pretty sure they e been wisely collecting true data for several years now, and probably have some really big storage files tucked away. So, maybe some day they’ll go at it from that perspective.
Don’t get me wrong here. Both ARM and OptionFinder are incredibly clever, sophisticated and unique tools. It’s just that, imo, using them together would not yield reliable, repeatable results.
[Edited by Jim Dean on 2/12/2022 8:03 AM]
|
|
Boris Lund
 Regular
  Posts: 59
Joined: 12/9/2021
User Profile |
I am glad you can see the potential :)
Yes I am aware of the "fake" historical data that OptionsTrader is using right now and why we cannot immediately start using ARM on it. I was told the same thing in support some time ago, but now I am asking again to get real historical data together with corresponding functions.
The options data would be better than any other data so we would get much better ARM training.
It is clear from spotgamma that the data they process is beyond what any individual can do. So can we team up with spotgamma to get in on their data? Or can we get some of that data ourselves?
I really think this is a road with much more potential than anything else, because we are looking at positions already placed with hedging levels.
|