The Subjective Function (Part 4)
Posted by Mark on October 10, 2012 at 07:31 | Last modified: October 1, 2012 06:03I’m in the process of trying to determine the subjective function, as defined in http://www.optionfanatic.com/2012/10/05/the-subjective-function-part-1. In http://www.optionfanatic.com/2012/10/09/the-subjective-function-part-3, I showed three inserts to help us in this determination. I will analyze those today.
Let me start by emphasizing the need to study relative differences in net profit, drawdown, and other details between the three inserts. If I focus on absolutes then I will be disappointed. The one and only goal of this study is to determine which I prefer most.
In looking at the equity curves, my first observation is that they are all upward sloping to the right. I should hope so. Each insert corresponds to the system that performed best on a different metric. Whichever curve I like the best corresponds to the metric I will use to evaluate systems. This will be the subjective function. Since each of these systems is the best in some respect, I am glad to see they are all profitable.
Equity curves #1 and #2 have many discernible plateau (horizontal) regions as opposed to #3, which seems to have more incremental changes. This corresponds to the total number of trades. Inserts #1, #2, and #3 have 57, 37, and 399 total trades, respectively. On this metric alone, I prefer a system that trades more because that means more opportunity for profit.
Drawdown curves #1 and #2 have a much smaller blue area than #3. This means #3 spends more time in drawdown and less time at new equity highs. I really like new equity highs because they make me feel like a successful trader. The broad market does not make all-time highs very often.
Equity curve #1 is lower now than it was around the beginning of 2009. This seems like a negative. However, each of these systems goes into drawdown periodically and curve #1 did hit a new high roughly 1+ centimeters ago (on my screen), which is only a couple fractions of a centimeter more than the other two. Curve #1 just happens to be in greater drawdown (-0.59%) than the other two but by no means it this drawdown remarkable. This observation may be disregarded.
I will continue this analysis in my next post.
Categories: System Development | Comments (2) | Permalink