Optionvue Update

I tested the new release. Seems wrong. Very different from the beta i tested. Already written to OV. Can you guys test and see if it's really a problem or the issue is on my end?

I suggest you save a copy of your current OV folder as backup before you upgrade.
 
My graphs are still way off unfortunately. Although EIOIO model also off so issue perhaps unrelated to curve fit.
In variable I also have to update asset files each time I open, otherwise the raw vertical skew data doesn´t load.

Posted by DavidF
 
Note from Len .....

I have found the problem. When first entering BackTrader mode, the program does not compute the MIV’s of the nearby options that are “brought back into existence” by going back in time. This was a new line of code that I added to prevent unnecessary MIV recalculations, and I did not realize that it would impact calculations that needed to be done at the time of first going into B.T. mode. (After you are in B.T. mode, if you change the date/time, the MIV calculations are always done then.)

I have removed that line and we will put out a new release soon. We sincerely apologize.

Users need to understand that this does not represent a “broken model”. It is merely MIV calculations that are not being done, and as a result, the skew curves for the nearby expirations are not being fitted. Everything should be working fine in real-time mode, and also in B.T. mode after you make any change to the date/time.

Len
 
Tested 7.83. It's very close to 7.64. This is based on back trader. Haven't had chance to compare live yet. But if it's like 7.82, then I think it is good enough in my opinion.

The data below is from my Jan 2015 trade. Bear in mind, the greeks are based on 200 butterflies. The difference in delta is within +/- 60 or about +/- 3 per tranche of 10 butterflies.

I have been trading live toggling between v7.8+ and v7.64 for more than a week now. Frankly, I find the difference in delta isn't large enough to change my adjustment decisions. With the small difference I'm seeing in v7.83 so far and if it holds in live trading on Monday, I'm not sure how others feel, but I think I'm ready to close this chapter and move on to something else. It has dragged on for a little too long.

upload_2016-1-2_2-59-25.png
 
Getting Len to change the Vol model back to its original state is a success for the community!

Any ideas on how we can get him to focus on modernizing the architecture of the solution to take advantage of today's computers?

I am thinking multi-core processor support, 64-bit, multi-monitor support, etc.

Posted by TheSpeculator152
 
tom said:
Getting Len to change the Vol model back to its original state is a success for the community!

Any ideas on how we can get him to focus on modernizing the architecture of the solution to take advantage of today's computers?

I am thinking multi-core processor support, 64-bit, multi-monitor support, etc.

Posted by TheSpeculator152
Click to expand...

Why not put in the specific request as a separate thread. Then add in poll for people to vote. If we get enough votes, maybe, we can get OV's attention.
 
Kevin Lee said:
Why not put in the specific request as a separate thread. Then add in poll for people to vote. If we get enough votes, maybe, we can get OV's attention.
Click to expand...

Good idea. Will do later today.

Posted by TheSpeculator152
 
Mine looking very solid now. I don´t have backtrader so am not sure why it´s suddenly working but the analyse graphs look ´normal´ both in live trading hours and otherwise. Also the large/medium/mall extents don´t throw everything off now. Finally the deltas for variable in relation to EIOIO are looking very much like they did before. A relief and good timing for a new year.

Posted by DavidF
 
In terms of the new version (7.83), I would add that it now incorporates the very excellent idea from Kevin about offering checkboxes for including multiple Recon identifiers. I can now mark my separate spreads in the T. Log as 'a', 'b', 'c', 'd', 'e', etc. and then quickly combine them in all sorts of variations in the Matrix and the Analysis tab by simply checking the boxes. This is a great step forward. Thanks to Kevin for the idea and the implementation.

(If you don't know what I mean, take a look at Kevin's video on ways to use the T. Log)

I should also add kudos to Len and the OV folks for paying attention to this new feature and incorporating it in the midst of what I am sure has been a stressful period.
 
JayH said:
In terms of the new version (7.83), I would add that it now incorporates the very excellent idea from Kevin about offering checkboxes for including multiple Recon identifiers. I can now mark my separate spreads in the T. Log as 'a', 'b', 'c', 'd', 'e', etc. and then quickly combine them in all sorts of variations in the Matrix and the Analysis tab by simply checking the boxes. This is a great step forward. Thanks to Kevin for the idea and the implementation.

(If you don't know what I mean, take a look at Kevin's video on ways to use the T. Log)

I should also add kudos to Len and the OV folks for paying attention to this new feature and incorporating it in the midst of what I am sure has been a stressful period.
Click to expand...

Thanx Jay. I didn't even noticed that. I'll definitely go play around with the new feature. :)
 
Not sure I share the enthusiasm about OV client service. They did fix a big problem, but the problem was of their own making and was due to their belief that they know better what they clients should have. For weeks they denied that the problem was real. I am glad we can now get back to trading instead of looking for work-around to the new "improved" model. But we are just back to the same unfriendly and tired interfaces, slow data update and delayed response to keyboard data entry.

Sorry to be so negative, but after years of managing a large IT and application development department I am sensitive to bad customer service.
 
Kevin Lee said:
Ron,

You are absolutely right. Depending on whether the skew is flat or steep, it'll sometimes be more advantageous to roll up (or down) the shorts and some other times to roll up (or down) the longs, although they provide the same amount of delta. Sometimes, looking at the skew curve, it can help us decide which exact strike to place the long and short options.
Click to expand...

@KevinLee,

would you please elaborate that a little bit more? maybe 2-3 examples would be great for us (newbies)
:)
[/URL] on how to choose adjustments depending on the vol skew. I think that would be very useful.

Thanks a lot,

David

Posted by dacamon
 
tom said:
@KevinLee,

would you please elaborate that a little bit more? maybe 2-3 examples would be great for us (newbies)
:)
[/URL] on how to choose adjustments depending on the vol skew. I think that would be very useful.

Thanks a lot,

David

Posted by dacamon
Click to expand...

Let me use this as an example - look at the IV curves below. The red circle on the curves denote ATM strikes. Market moved from orange line to blue line - ie market moved up. IV dropped. Notice that as IV dropped, higher strike IVs drop more than the lower strike IVs. Thus making the IV skew steeper. Take note the magnitude of drop is especially steep above the ATM strike. This is a common phenomenon. Not 100% but you can bet on it.

Knowing this behavior, when market IV is relatively high and the IV skew relatively flat, it would be advantageous to buy lower strike options and sell higher strike options. Example, after a market sell off, IV is relatively high and IV skew relatively flat and you need to adjust. Assume you need to add more positive delta. There are multiple ways to get the same amount of delta. There are pros and cons of each method and different trade offs in terms of greeks. But from a skew perspective, it would be advantageous to place the short strike above ATM and the long strike at or slightly below ATM. As market stabilizes and IV falls, the short strike will experience a much larger IV crush than then long strike.

Make sense?

upload_2016-1-19_5-38-43.png
 
Kevin Lee said:
Let me use this as an example - look at the IV curves below. The red circle on the curves denote ATM strikes. Market moved from orange line to blue line - ie market moved up. IV dropped. Notice that as IV dropped, higher strike IVs drop more than the lower strike IVs. Thus making the IV skew steeper. Take note the magnitude of drop is especially steep above the ATM strike. This is a common phenomenon. Not 100% but you can bet on it.

Knowing this behavior, when market IV is relatively high and the IV skew relatively flat, it would be advantageous to buy lower strike options and sell higher strike options. Example, after a market sell off, IV is relatively high and IV skew relatively flat and you need to adjust. Assume you need to add more positive delta. There are multiple ways to get the same amount of delta. There are pros and cons of each method and different trade offs in terms of greeks. But from a skew perspective, it would be advantageous to place the short strike above ATM and the long strike at or slightly below ATM. As market stabilizes and IV falls, the short strike will experience a much larger IV crush than then long strike.

Make sense?

View attachment 2518
Click to expand...

Yeah, that makes a lot of sense. Thanks a lot for all this information. It really helps a lot. A couple of additional comments:

- When trading small size, it is not easy to place the longs and shorts that far one from the other, like 50 or 60 points (because that will give you more deltas than needed)
- So, mainly, I have to decide whether I place the spread ITM, ATM or OTM (usually 20-30 points wide). I understand that the vol skew can help but only a little in this case, and mainly when you place your spread ITM. So , in that situation and based on your experience, do you think it makes lot of a difference in the PNL?
- Last: Do you use any specific software to analyze the Vol skew?

Thanks Kevin!

Posted by dacamon
 
I used Kevin's IV skew graph above to discuss the topic in more depth during today's Trading Group 2 meeting (Jan 19, 2016).
Please review that if you'd like a deeper understanding of how skew works, in terms of trade design and adjustments.
My section starts from the 13 min mark.

 
You must log in or register to reply here.
Top
Contact Us