tag:blogger.com,1999:blog-5108179989881744725.post1315558439378472783..comments2023-12-13T04:12:08.163-08:00Comments on Trading with Python: Short VXX strategysjevhttp://www.blogger.com/profile/17452562180989360928noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-5108179989881744725.post-14386082788789965132016-06-17T11:38:23.562-07:002016-06-17T11:38:23.562-07:00Hey Jev, long time. I too have been messing with V...Hey Jev, long time. I too have been messing with VXX and XIV although a lot less quantitative. Long XIV and long VXX calls (some long VXX puts that have worked out with mixed results). Looks like I'll have to re-familiarize myself with my TradingwithPython course.Unknownhttps://www.blogger.com/profile/06688547307798884714noreply@blogger.comtag:blogger.com,1999:blog-5108179989881744725.post-42975180532945412662013-08-19T14:26:41.287-07:002013-08-19T14:26:41.287-07:00The borrowing costs are very reasonable (couple pe...The borrowing costs are very reasonable (couple percent/year) given the yearly loss of value in VXX .sjevhttps://www.blogger.com/profile/17452562180989360928noreply@blogger.comtag:blogger.com,1999:blog-5108179989881744725.post-82294891573934231192013-08-19T12:45:40.718-07:002013-08-19T12:45:40.718-07:00Interesting idea, what sort of borrow costs have y...Interesting idea, what sort of borrow costs have you encountered when shorting VXX? I would suspect that they are quite high for leveraged/volatility ETFs ?one_fell_swoophttps://www.blogger.com/profile/04415486703354218677noreply@blogger.comtag:blogger.com,1999:blog-5108179989881744725.post-90561398643696111552013-08-19T02:47:36.457-07:002013-08-19T02:47:36.457-07:00This post only gives an overview of building and t...This post only gives an overview of building and testing the strategy.I plan on explaining this step-by-step (including getting the data from CBOE and curve fitting) in the TradingWithPython course material. <br /><br />Regarding VXV=f(VIX), I know (from previous testing) that the relationship is pretty stable through the years, so I am not afraid of look-ahead bias. The last 250 days are plotted in the first chart in red, just to make sure that the relationship is roughly the same as before.<br />sjevhttps://www.blogger.com/profile/17452562180989360928noreply@blogger.comtag:blogger.com,1999:blog-5108179989881744725.post-45205981825475888272013-08-19T02:27:30.585-07:002013-08-19T02:27:30.585-07:00This post is quite heavy to digest. How about spli...This post is quite heavy to digest. How about splitting it into pieces? It would be great to get more details on your 2011 trades. You may show some charts with dots where you get in, out and switch from one ETN to another. That's a lively experience to share!<br /><br />Also, defining the underlying of VIX and VXV may explain better the relationship between the spread and the contango or backwardation, from a didactic point of view.<br /><br />Your strategy is amazing and supplementary to Michael Stokes' blog (http://marketsci.wordpress.com/). My only critic is that you use the function VXV = f(VIX) in your backtest but the function is known only ex post. You may amend your process to walk-forward estimate the function f.<br /><br />I love your posts and expect to read more in the near future. Congrats again!Yaba Qihttps://www.blogger.com/profile/03005683436581630130noreply@blogger.com