5 Minute Read
04 December 2020
By Conor Naylor
Aikido Finance is delighted to announce the launch of our Premium Investment Strategies, and they’re free for all Aikido early access users! Returning 15-20% per year, these long term investment strategies have outperformed the market, at lower risk over the past 20+ years.
At a high level, a quantitative strategy is a set of filters which are applied to a database so that it selects a number of stocks to invest in. A “good” strategy is one which provides a historically high return, at a low risk.
Building a basic investment strategy follows a simple approach:
Next we add our static filters to the remaining stocks. We can choose any metric from any of the high level factors.
At this point we could be down to just 200-300 stocks, depending on how broad the filters are in the previous step.
Next we apply a ranking/ordering to the remaining stocks. Again, we can pick any metric here. The ranking operator has a big effect on the performance of a strategy as it can be the attribute by which most weight is given.
For example, we can rank our stocks by their descending 6 month price momentum, ie. those with the largest stock price increase over the past 6 months.
The number of stocks in a portfolio can also have a big impact on the performance of the portfolio. A more concentrated portfolio of 5-10 stocks will likely produce higher returns but with greater volatility. We usually like a portfolio with 20-25 companies in it.
Et voilà, you have created a quantitative strategy which picks undervalued, high quality stocks with high momentum.
Our premium strategies take a slightly different approach to the filtration process. The Aikido Team has been busy innovating on our strategy curation process, and the results are spectacular. We’ve read the research and incorporated new features into our application, based off of our findings. Our premium strategies utilise a new stock selection flow which reduce risk while improving return.
An issue we discovered during our research is that static filters do not account for certain changes in the economy and therefore changes in the underlying fundamental data of each stock. For example, filtering for stocks with a price / earnings ratio less than 10 gives a very different result now than it did in 2008. We looked for a more dynamic approach to solve this problem. Our premium strategies can now apply a sequential filtering process. This filtering process allows us to build a dynamic factor model using multiple metrics. The sequence by which the metrics are used, determines the concentration and exposure of each metric in the final portfolio.
Our new approach allows us to rank and filter stocks multiple times in one strategy. Take for example our Sequential Factor Blend strategy. First, we select the top 100 most efficient stocks, measured by Return on Equity. Then, we take the top 50 of these stocks, ranked by ascending Price / Earnings ratio. Finally, these stocks are ranked by their 3-month price momentum, and the top 10 are chosen. This strategy has average annual returns of 14.62% per year, backtested since 2000.
We have included a new feature called the Trend Follower. The trend follower is based on research conducted by Gary Antonacci into Dual Momentum. It is an indicator which provides a Buy/Sell signal during the rebalance process, depending on the momentum of the underlying benchmark. An example trend follower would check if the S&P 500 is positive, if so, rebalance as usual, otherwise sell your positions and hold cash. The trend follower greatly reduces volatility as it prevents users from holding stocks during downturns in the market. Take look at a comparison between two premium strategies: Smart Quality & Top Quality and High Momentum. The performance of each is impressive, but the inclusion of the trend follower in the Smart Quality strategy improves the risk adjusted returns, volatility and max drawdown for just a small reduction in the average annual return.
These more performant investment strategies are easy to implement and easy to understand. Check them out here.