tag:blogger.com,1999:blog-261139923818144971.post107655729376816275..comments2024-03-27T07:58:49.946+00:00Comments on This Blog is Systematic: Computers vs Humans - considering the medianRob Carverhttp://www.blogger.com/profile/10175885372013572770noreply@blogger.comBlogger11125tag:blogger.com,1999:blog-261139923818144971.post-57547056486033008022016-04-12T10:47:55.961+01:002016-04-12T10:47:55.961+01:00Ok. So it would definitely cut my expected return....Ok. So it would definitely cut my expected return. Thanks for the clarity. Much appreciated.<br /><br />Cracks and spread diversification is an interesting idea. But they can still experience periods where they are quite flat price dependent I find. So probably keeping it as a smaller separate strategy set within the portfolio is logical.<br /><br />A forecast scoring system for both fund/tech sounds like an interesting route. I guess I need to read your book further to discover the full meaning and how to apply a forecast scoring model. <br /><br />Thanks very much. <br />SimonSimon Bhttps://www.blogger.com/profile/18065446526614832707noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-88901354919848106662016-04-11T15:46:17.209+01:002016-04-11T15:46:17.209+01:00Hi Simon
It's true that for basic technical t...Hi Simon<br /><br />It's true that for basic technical trading much of the benefit comes from diversifying across asset classes. Not doing so will cut your expected returns by around a factor of 2.<br /><br />There are certain kinds of niche technical models that one can run in energy markets. For example mean variance of the crack spread. Calendar spreads and butterflies in crude oil also show interesting behaviour. These are all relative value strategies, so negative skew / high leverage, and you should be wary of putting too much into your portfolio.<br /><br />As you say the other advantage you might have is developing a system that combines fundamentals and technicals. For the former you might decide it's too much work to make it fully systematic, but come up with some kind of forecast score. You can then combine that with the forecast from technical models.<br />Rob Carverhttps://www.blogger.com/profile/10175885372013572770noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-44021788266731127112016-04-11T15:38:44.274+01:002016-04-11T15:38:44.274+01:00Hi Rob - first off, great blog resource here. I...Hi Rob - first off, great blog resource here. I'm a big fan. Currently reading your book. Very interesting.<br /><br />I have a question. DO you have any experience with systematic trading specific to niche or a single set of markets ? (e.g. crude markets)<br /><br />I'm a junior paper trader in a physical commodity house. I trade Brent, Gasoil, Gasoline, Heating oil. I've been playing about with and testing various technical ideas like composites of moving average crossovers with some varying/interesting results. <br /><br />I was wandering what advice you might have with regards to applying systematic or semi-automated trading to a much smaller universe of markets ? Is it even advisable ? Would you need a big library of different technical or fundamental strategies to compensate for the lack of market diversification ? Or is that a dangerous game ? <br /><br />I cant prop in FX, STIRs, Equities etc.. just energy markets you see. Interested in your thoughts. Thks.Simon Bhttps://www.blogger.com/profile/18065446526614832707noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-77042271297121652372016-03-09T12:04:41.547+00:002016-03-09T12:04:41.547+00:00OK that's interesting - so you could do a fore...OK that's interesting - so you could do a forecast depending on where you are within a 26 week price channel with the EMA being the midpoint and the channel defining the forecast extremes for example (or use a Bollinger Band) - I'll try that - thanksCJhttps://www.blogger.com/profile/06817263570026256611noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-67409047822635384172016-03-09T10:59:39.504+00:002016-03-09T10:59:39.504+00:00You're right, it's not possible. For this ...You're right, it's not possible. For this reason I tend to avoid strategies like this. Or try and adapt them so they are continous in nature.<br /><br />So, I use a stochastic as my "breakout" signal (price - mid range of price / range of price). This means that I'll be fully long at the point of the breakout, but I'll have started going long the moment the price is in the top half of the recent historical range.Rob Carverhttps://www.blogger.com/profile/10175885372013572770noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-84936760338183587092016-03-09T10:52:16.869+00:002016-03-09T10:52:16.869+00:00Hi Rob
Really really appreciating your work and my...Hi Rob<br />Really really appreciating your work and my comment is not related to the above post and you may have answered the question before so apologies in advance. The question I have is, do you have a simple method of converting breakout strategies to a forecast. Obviously some things like MA crossovers easily translate to a forecast but as far as I can see I only translate a breakout into + 20 go long, - 20 go short. Any thoughts?<br />Many thanks ..CJhttps://www.blogger.com/profile/06817263570026256611noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-36893829738147815912016-02-21T17:33:50.869+00:002016-02-21T17:33:50.869+00:00Donald,
I like the sound of that book. Must add it...Donald,<br />I like the sound of that book. Must add it to my reading list.Rob Carverhttps://www.blogger.com/profile/10175885372013572770noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-85914247418238399482016-02-14T21:05:08.959+00:002016-02-14T21:05:08.959+00:00Rob, I would like to second Matt's comment...g...Rob, I would like to second Matt's comment...great stuff. I just read this post while half way through http://www.amazon.com/Your-Money-Brain-Science-Neuroeconomics/dp/0743276698 ... it really is true what old Ritholtz says: "We are monkeys with shoes"!!<br /><br />Cheers, Donald (down here in NZ)Unknownhttps://www.blogger.com/profile/08980937836132071634noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-24747337757102814142016-02-07T17:59:34.466+00:002016-02-07T17:59:34.466+00:00I will only say that to paraphrase Justin Mamis. ...I will only say that to paraphrase Justin Mamis. In the endcomputers are programed by humans, therefore they eventually express human error and frailities Mikehttps://www.blogger.com/profile/12803900184577731114noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-71612627967319093192016-01-30T09:45:38.416+00:002016-01-30T09:45:38.416+00:00Thanks Matt I really appreciate that.Thanks Matt I really appreciate that.Rob Carverhttps://www.blogger.com/profile/10175885372013572770noreply@blogger.comtag:blogger.com,1999:blog-261139923818144971.post-57431575058418928622016-01-30T07:54:00.784+00:002016-01-30T07:54:00.784+00:00There is far too little humor in finance. Thanks f...There is far too little humor in finance. Thanks for mixing numbers and fun in just the right ratio.Matt Haines Photographyhttps://www.blogger.com/profile/13895187802989669254noreply@blogger.com