The XIV model adjusts to the available market data by using the opposite of the daily return of the VXX and taking to account trading costs. So if VXX moves x%, XIV moves -x% that same day. That is not the same as a mathematical inverse 1/(1-x). This causes that it does not follow the same path of the VXX. I mean if in a few days VXX goes from A to B and then back to A, XIV will not go from A’ to B’ and back to A’.

It seems to me that models are as good or bad as their assumptions. The model I used is explained by following the XIV link on the menu. It adjusts quite well to the available market data and projects where that data is not available. It is based on the VXX price which is based on the VIX futures price. That means that if you know the VIX futures you can have a good idea of the XIV price. But it is to me impossible to predict what the VIX futures will do, therefore also impossible to predict what the XIV will do. At most I could say something like: if the VIX futures behave like this then the XIV price should behave like that. But even that can change if the way that the XIV is managed changes.

I do not know if we can expect the same level of behavior in XIV/VXX or VIX in the future. I’m pretty sure there will be violent moves sometimes but I do not think they are predictable. We could have 10 or 20 years of high or low volatility. Or 5 years of altering high and low extremes. Anything could happen which would make the past 10 years’ behavior look different than the next 10.

Feel free to mail me to discuss our common interests, you can write me via the contact menu and I’ll reply with my e-mail.

Best wishes !
jrv

By making a chart of the B column you can easily see how volatile this ETF is. http://img819.imageshack.us/img819/3218/hypotheticalxiv.png

There was an hypothetical maximum of $36.41 on 2/22/2007, and after that a minimum of $2.13 on 11/20/2008. That’s really sick, nearly a 95% loss in 21 months. But in the other side we can also see that this ETF would had increased in 2281 days from 4.9 to 21.27. That’s an increase of more than 0.06% per trading day or about 17% per year. JRV, do you think that we should expect that kind of moves and profits in the long term?

I have read many studies that calculate about a 4%-5% monthly decrease in VXX, shouldn’t we, therefore, expect an increase of that amount in XIV/SVXY minus VXX and XIV/SVXY costs? That’s much more than the previously calculated 17% per year.

I have made a simulation of an ETF that does every day the opposite of VXX (of your simulation of VXX since 2004) and my results are quite different. I know that I should take into account expenses, but the difference shouldn’t be so big. In this case, in that 2281 days period its value is multiplied by nearly 13 instead of only 4. I can send you that spreadsheet if you want.

You are being way too hard on yourself. The index number is a 100% invested number with very high potential volatility. Your portfolio not only focuses on companies with strong balance sheets, but also contains hedging to further decrease volatility. In volatility-adjusted terms, you are doing an outstanding job. I aspire for these kinds of volatility-adjusted returns someday, and I will be thrilled if I can beat the market indexes with much lower risk like you do.

It has been so long since i have sent you any comments, currently i am now in NY, as i think i told you i am a Doctor and i just love the stocks but i am not that good at it.

The thing is that i read this terrible news in The New York times that i wanted to share with you since i trust your opinion so much. http://www.nytimes.com/2013/03/31/opinion/sunday/sundown-in-america.html?pagewanted=1&_r=2&ref=general&src=me

By the way i have many books of Dale Carnegie, i got them through my dad who already passed away and i am really happy that you like them too.

Cheers de tu amigo Pablo.