Weekly Algorithm Review: 01/28/2023 to 02/03/2023
Algorithm Performance This Week
Overall Market: +2.83%
Base Algorithm: +1.59%
Variable Market Neutral: +0.28%
Long Term Portfolio: -0.4%
Variable Sector Neutral: -0.46%
Market Neutral: -0.75%
Sector Neutral: -1.12%
Performance Summary - Technicals
On the technicals side: no notes. Every technical algorithm performed as we want them. The base algorithm outperformed the long term portfolio by more than a percent. The variable hedgers fell between their respective hedger, and the base algorithm. Market neutral and sector neutral underperformed substantially, but this is to be expected when the market performs that well. I would have liked to see them more positive, but one week of poor hedging isn’t a red flag. I’ll be watching their performance from this point on.
Similar to last week, a major reason for the base algorithm’s strong performance was it dodging a few severe losers. Given that our systems are all based on machine learning, they can be somewhat black-box at times. That is: while it’s easy to see what they’re choosing, it’s much harder to see exactly why. It is poor form to extrapolate from 2 data points, but I suspect that our system has found a set of signals that precede major drops (or a disproportionate chance thereof) and avoids them. It’s possible that, in its training data, this is the best strategy it found. Given that our lookback is 1 year, and the nightmare 2022 was for trading, I wouldn’t be surprised if this was the case.
I’m interested to see if this trend continues.
Performance Summary - Fundamentals
I’ll be blunt. Our long term portfolio is continuing to underperform. For this reason, we are beginning work on a more bullish portfolio. We hope to roll this out next week.
Misc. Data For The Week
