By: Jeffrey Cohen, Investment Advisor Representative
Well, it's late and the run did not finish. We added back some extra solvers including the particle swarm optimizer, the simulated bifurcator, the 'fast' genetic algorithm, the gradient descent model, the D-Wave simulated annealer (not quantum), the TABU multi-start solver, and even our own custom coded simulated annealer. This model takes so many hours to run, and crunches the data so many ways, that we sure hope we find the best portfolio.
That portfolio will have just the right amount of zig and zag in the prices, so that the price variance of the overall portfolio is the lowest given the highest expected return of that evenly weighted portfolio. Put another way, it finds the portfolio with the lowest risk and the highest return that it can.
We could run it faster, but when the market returns are so low (like they are now), risk matter more than return and it is very hard to find lower risk portfolios.
That said...we will just keep on waiting and wake up early and share our results with you, our faithful readers.
One more thing...there is a stock we like tonight. It is a retailer with management changes, a declining business forecast (due to recession fears no less), and assortment balancing problems. However, it is conservatively run, a great household name brand, a solid balance sheet, and a history of paying dividends and taking care of shareholders. It is an icon of the retail industry with deep roots and solid foundations. It will be a great addition to our 'long portfolio' when we buy some in the morning.
Call me for the stock. I promise to tell you. Just did DD on it. If the price hits the right entry point, we will be in. Whether you buy a run or a service from us is up to you. However, you need to read our brochure first (or at least download it) before you buy. FINRA/SEC wanted me to tell you that.
Just checked one more time...needs a little more time to run and crunch those portfolios. With 2,384 stocks with $100M equity market capitalization, positive net income, and liquid stocks that pass our stringent data validation process, there are many permutations to search. If I get my math right, it is on the order of 10 ^ 750 possible solutions (give or take a few trillion combinations).
The next morning...
The number crunching is done. A few insights appear. There are five leptokurtic stocks with low variance. These are familiar names to us, but are not necessarily stocks we can act on.
['AQN', 'HCIC', 'PNM', 'SLF', 'WMPN'] 5
The early solvers in our process found small portfolios. I think those are easiest to find by our solvers. After the runs are complete, the best portfolios range from 6 to 10 stocks, equally held. These are a combination of high BETA stocks and stocks in the technology space.
Here are the 6 to 10 stock portfolios:
-0.0009 ['AOSL', 'CLFD', 'SI', 'SITM', 'SQ', 'TTD'] 6
-0.0009 ['AOSL', 'CLFD', 'DDS', 'SI', 'SITM', 'SQ', 'TTD'] 7
-0.0009 ['AOSL', 'BMBL', 'CLFD', 'DDS', 'SI', 'SITM', 'SQ', 'TTD'] 8
-0.0008 ['AOSL', 'BMBL', 'CLFD', 'DDS', 'LC', 'SI', 'SITM', 'SQ', 'TTD'] 9
-0.0008 ['ACLS', 'AOSL', 'BMBL', 'CLFD', 'DDS', 'SI', 'SITM', 'SQ', 'TTD'] 9
-0.0008 ['ACLS', 'AOSL', 'BMBL', 'CLFD', 'DDS', 'NVDA', 'SI', 'SITM', 'SQ', 'TTD'] 10
We found 50% better solutions by letting our solvers run longer.
Only two stocks make the cut today because the indices score better today. The SPY (S&P 500 Index ETF) was the 13th most efficient individual stock.
BX (Blackstone Inc)
CG (The Carlysle Group)
Good luck in the markets today.