Thanks JackRab. But note the writer didnt incorporate those fees because he was NOT trying to mimic VXX. His test was shorting VIX and just holding the front month futures as a hedge. So at most that is missing commission wise on the hedge side is one trade a month.
@Saltynuts, I don't think his study is correct.... The fact that VXX usually drops is due to the fact that we usually see contango in VIX futures, with the sharpest decline is the nearer we are to expiration. So, the front month VIX future drop at a faster rate than the back month future. Which would mean, when he buys and holds the front month VIX future... it would decline in value more than the VXX, since VXX is composed of a combination of front and back month futures... depending on if we are near expiry or not. So with 30 days to go, VXX is full in front month, zero in back month 25 days to go, VXX is 25/30 in front month, 5/30 in back month... etcetc. VXX gradually moves out of the faster losing future... so it should lose less than just holding the front month. I don't know exactly what the article writer has done... but it doesn't seem right to me.
Yes, it is VERY weird JackRab. On another threads someone (maverick?) mentioned the long volatility funds "locking in losses" every day. That does not seem quite right to me, as if VIX goes up they are locking in gains, but it could have something to do with the rollover of course. Just weird.