(Restricted access - clients only)
Unicorn is the suite of systematic strategies traded within KM cube’s infrastructure through an AMC or mandate. The advantages of systematic strategies are well known in the literature that offer superior risk controls, avoid behavior biases and offer low correlation to traditional portfolios. There are four strategies currently deployed: Volatility, Momentum, Value, and Traditional. In the section below we present the summary of improvements applied
Summary of improvements
Volatility strategy
The legacy strategy uses three factors to generate signals, volatility premium, slope, and gamma exposure. As the recent data show, curve and gamma exposure are less effective and for this reason, we decided to drop them. The results are marginally different, so we prefer simplicity in order to reduce degrees of freedom, and thus the likelihood of fitting the model to the data.
Value strategy
The value methodology has changed and we have shifted to single stocks instead of ETFs. Our methodology relies on fundamental analysis (using Piotroski) of thousands of equities in the US, Europe, Japan, Australia, and Canada. The top 10 stocks are traded on an equal basis every month with a trailing stop loss of 10% applied at the beginning of each month.
Model ranking
We are changing the way unicorn allocates capital between strategies. Instead of ranking based on past performance, we allocate capital equally based on risk. This brings better stability to the model as it reduces the overreliance on one strategy.
Traditional model: back to bonds
The 60/40 portfolio represents a simple yet powerful allocation of bonds and stocks. In its simplest form a portfolio is invested 60% in equities and 40% in stocks, rebalanced quarterly. However, this approach is too simplistic as it relies on a the premise of low correlation between stocks and bonds a rather inconsistent relationship.
To overcome these problems, the “traditional” model relies on trend indicators to allocate capital on stocks or bonds and is applied only when a trend is present. This is a simple yet powerful way to mitigate the correlation effect. During the last two years, an additional improvement was to substitute the bond exposure with gold as the ultra low interest rate environment was offering very limited upside for bonds.
Following the substantial sell-off in bonds this year, a switch back to bonds offers renewed potential.
Effective 2022-11-01 the 40% allocation will be US government bonds.
The next two pages show backtesting results and a comparison of profitability.
Backtesting results
Before
After
