You Analyze, I Analyze, We Analyze, But we disagree (Part 3)
May 30, 2010 1 Comment
This will be last part of my series on analysis and I will be touching on noteworthy ideas that i wish to investigate further through this blog and also on my free time. But before I start I want to make some minor additions to my prior posts. Firstly, I would want to add that qualitative analysis is also actually a model, it just that its a mental model and hence like any other model cannot be all encompassing. In addition another article form the NYtimes once again highlights that even irrelevant information can effect our decision making. Now back to noteworthy ideas.
Firstly a study on momentum. Momentum basically states that strong stocks will stay strong and continue higher. AR has also summed up strength of momentum strategy how to mix it with other strategies. Meanwhile CSS analytics has also provided a nice starting point on how to think and model momentum otherwise also known as relative strength. CXO advisory adds to this by showing testing out the different variables of momentum here, here and here. But above all we have a great blog really focusing on systematic relative strength.
Secondly a study on machine learning (adaptive algorithm)l. Well this is going to be challenging, but after coming across this article on CSS analytics, it seems like this idea of machine learning is a great way of making any strategy robust. The idea behind machine learning is basically to get an algorithm that filters out the market data based on lagged periods behind the current price and volume other other variables. Through this filters we then determine what kind of strategy to implement now. Here is the link to the paper done machine learning whilst the complementary video is found here. Another useful article is also found here where the writer does will in linking EMH, AMH and Machine Learning algos together.
Thirdly, studies on the 200 MA. This is one of the robust trading signals ever, especially on stock indices. Even some fundamental traders also taken note of the importance of this moving average. There was also a nice study done by CXO advisory on comparing the 200 day moving avg, 40 wk moving avg and 10 month moving avg and they found that
In summary, evidence from simple tests indicates that a 10-month SMA outperforms 40-week and 200-day SMAs under reasonably realistic trading assumptions for the broad value-weighted U.S. equity market over the past 17 years, but the source of outperformance may be luck.
Last but not least a study on the VIX, Bill Luby who has a great understanding of the VIX (and index reflecting the implied volatility of S&P index options), has shown that VIX tends to spike upwards and points out the reason here. This basically means that fear gets into the market in a shorter amount of time, causing volatility to rise however humans do not get immediately rational and spike back into neutral territory.
But most noteworthy of all is to note that investing or trading at its core is supposed to be a simple process. Over complication reduces the robustness of the method.