Research

Automatic Support/Resistance using ML

I think an interesting application of ML could be generating the ‘features’ for inclusion in a trading algorithm, converting non-numerical data into numerical. For example, converting market sentiment or satellite imagery to count cars in a retailer’s parking lot. More examples here. I had written previously about Machine Learning for trading strategy design here. Thanks to code from @johnromero via @quantocracy I wrote a quick script to draw support and resistance lines automatically using machine learning on…

Read More

Machine Learning & SciKit Learn

I made the point to someone the other day that technology and coding is getting easier and easier to accomplish. I don’t think I would have been able to perform ‘machine learning’ five years ago but with the resources available today (Python, SciKit Learn, and pages upon pages of StackOverflow) even someone like me can fit a model and build ML algorithms. Machine Learning is also ridiculously “easy”. It’s literally 4 lines of code. It…


Speculation in a Path Dependent World

I’m happy to publish my first paper on SSRN entitled, “Speculation in a Path Dependent World”  I found many otherwise talented managers entering a multi-manager platform (assigned capital with strict risk limitations) having a difficult time transitioning. The paper is my simple attempt to suggest an alternative framework for new or potential managers dealing with drawdown risk. Any and all feedback, comments, advice, diatribes are greatly appreciated!  


2015 Year in Review

2015 is in the books and it had it’s moments, both good and bad. I had the opportunity this year to travel to 6 continents including an amazing trip to the edge of the world, Antarctica. Unfortunately, we had some health issues in the family as well but we are optimistic in that department for 2016. Leaving the firm last year offered me an amazing opportunity to travel extensively which is an opportunity very few…


VIX ETP Performance in 2014

As pointed out by my friend @stkbullgod the VIX ETPs had an interesting year. Check it out: One would expect SVXY to be UP around 25% given where VXX is but that is not the case. There is a 36% difference! The VXX borrow rate was ~5% on last check so let’s say 30%. Similar results with XIV (I focus on SVXY as it is an ETF with options as opposed to XIV which is…


No Picture

Short SPX Calls

Many of the strategies I utilize or investigate tend to be correlated with short volatility strategies or perform in ‘stable’ markets and performance tends to deteriorate in risk-off phases. Additionally many long-short or qualitative managers short an index or ETF like SPY or IWM’s when they feel it necessary to have something on the other side with lack of an alpha short. In looking to hedge some of these strategies I always wanted to consider…


No Picture

Volatility is Back

Volatility has really made a comeback but it pales to the volatility seen in Japan. Here is a look at the Nikkei volatility index versus VIX: I don’t look often at relative comparisons from international asset classes on a daily basis. Perhaps I should as it seems like quite an attractive relative value trade. US investors could sell EWJ vol and buy SPX vol. My bias has always been to short volatility on pops and…


No Picture

More on Breadth Models

I’ve experimented with a few breadth models in the past. My hypothesis is that markets have become so efficient and correlated, that all (profitable) strategies break down into long vol or short vol profiles and at the end of the day, your profit will STILL be determined by the movement in the underlying. While utilizing an options structure might benefit or alter the structure of your payout, the extreme inefficiencies that once existed are probably…


No Picture

Follow Up to Op-Ex and Market Timing Model

The Op-Ex indicator started off slow since I posted with 1 break-even trade, 1 stop out, but the last one has been a nice winner: Keep in mind I haven’t really done much work on optimizing the trailing stop exit, there’s probably opportunity there for improvement.   The Market Timing Model also started off slow with some big whipsaws but looks like performance is improving:  


No Picture

Ops Ex

Been awhile since I posted anything, I’ve been to busy watching European theater as everyone else has been. But I did find an interesting pattern worth sharing; there has been a strong tendency for the short term trend to reverse after options expiration. I don’t have a very plausible explanation for why but I’m sure one can be rationalized if need be. In any case, I applied a simple stop loss and trailing stop strategy…