Chicago Quantum Net Score, Long, Some Considerations
Each time we run our Chicago Quantum Net Score Long (or UP) Analysis, we 'crunch the data' to identify the optimal long portfolio that maximizes historical risk minus expected return, hence the 'net score.' The expected return of the portfolio is an equally weighted portfolio where each stock has its own expected return based on the one-year BETA of each stock and the overall one-year market return. The historical risk of the portfolio is the variance of that equally weighted portfolio with the variance of each stock calculated over the prior year.
What makes this interesting is that you can get a stock portfolio, or identify individual stocks, that can really scream higher if the market rises, but they show lower historical variance so maybe they are less risky and have smaller drawdowns. At least, that is the theory.
What makes this computationally hard is that we are looking at portfolios, and not individual stocks. We benefit from diversification in the math for the long portfolio, because the model takes into account the correlations and covariance between stocks in the portfolio.
On the 'down' run, diversification is generally not helpful for more than 3 or 4 stocks. Your 'short' list should be small, because diversification eliminates much of the risk of decline in short portfolios.
Our model uses a collection of published search methods to find the best, and worst portfolios depending on the service purchased.
A few points of explanation:
The model does a very good job of finding what seems like a well-optimized portfolio, or set of portfolios in order of optimization, when historical market risk is lower. In that case, when historical market price volatility is lower, the balance of market power is leaning towards maximizing expected returns. When the model is focused on maximizing expected returns because historical risk is lower, it finds well optimized portfolios more quickly.
The converse is true as well. When historical market risk is elevated, the model has to work much harder to find optimized portfolios that minimize historical risk. It can seemingly look forever and keep finding slightly better alternative portfolios if we let it, or search in less efficient ways.
We are currently in a framework where historical risk is lower, historical market returns and expected returns are higher, and the model finds optimized portfolios quickly and efficiently.
We provide our clients who buy our CQNS Long Analysis the best 50 optimized portfolios in order of their Chicago Quantum Net Score. This allows them to trade off the benefits and costs of substituting different stocks, holding different combinations of stocks, or maybe avoiding certain market sectors.
We provide our clients with a spreadsheet of all stocks that pass data validation and are included in our data analysis, and that spreadsheet includes the CQNS score for each ticker. Clients can also look at the worst individual stocks to either avoid or short that list (if they want to act on one stock at a time).
What makes this interesting is that you can get a stock portfolio, or identify individual stocks, that can really scream higher if the market rises, but they show lower historical variance so maybe they are less risky and have smaller drawdowns. At least, that is the theory.
What makes this computationally hard is that we are looking at portfolios, and not individual stocks. We benefit from diversification in the math for the long portfolio, because the model takes into account the correlations and covariance between stocks in the portfolio.
On the 'down' run, diversification is generally not helpful for more than 3 or 4 stocks. Your 'short' list should be small, because diversification eliminates much of the risk of decline in short portfolios.
Our model uses a collection of published search methods to find the best, and worst portfolios depending on the service purchased.
A few points of explanation:
- Our CQNS stock portfolios are evenly weighted, so if you want to hold 10 stocks, each is 10% of the portfolio.
- We include in our search every stock that trades that day and that passes our data validation criteria for liquidity, absence of anomalies, and sometimes a market cap or stock price floor. On January 28, 2024, the analysis included 3,266 U.S. Common Stocks
- The current results of the model are picking portfolios that are expected to do well if the overall U.S. Stock Market does well. It is expected to do poorly (e.g., lose value) when the overall U.S. stock market does poorly. It reinforces the direction of the market.
- When we established our separately managed accounts based on the Chicago Quantum Net Score, we hedge those accounts (per client request).
The model does a very good job of finding what seems like a well-optimized portfolio, or set of portfolios in order of optimization, when historical market risk is lower. In that case, when historical market price volatility is lower, the balance of market power is leaning towards maximizing expected returns. When the model is focused on maximizing expected returns because historical risk is lower, it finds well optimized portfolios more quickly.
The converse is true as well. When historical market risk is elevated, the model has to work much harder to find optimized portfolios that minimize historical risk. It can seemingly look forever and keep finding slightly better alternative portfolios if we let it, or search in less efficient ways.
We are currently in a framework where historical risk is lower, historical market returns and expected returns are higher, and the model finds optimized portfolios quickly and efficiently.
We provide our clients who buy our CQNS Long Analysis the best 50 optimized portfolios in order of their Chicago Quantum Net Score. This allows them to trade off the benefits and costs of substituting different stocks, holding different combinations of stocks, or maybe avoiding certain market sectors.
We provide our clients with a spreadsheet of all stocks that pass data validation and are included in our data analysis, and that spreadsheet includes the CQNS score for each ticker. Clients can also look at the worst individual stocks to either avoid or short that list (if they want to act on one stock at a time).
Contact us to learn more. You may purchase a sample CQNS report for $5.00, or a recent, full CQNS report for $50.00.