I agree that a large model to cover the market in it's entirety is untenable due to mass variants. But what about a handful of stocks?
When I day traded I thought of it as cars on a highway that I was getting into, while in motion, for a period of time/distance. I noticed that I also was looking at stocks in other lanes, choosing my choices based on their actions. So, and I have this dream a lot, it became a matter of driving my car from above it, like a hundred feet, the attempts to micro control I found disrupted the car itself. The factor I hadn't accounted for was how I changed the car in my own view by getting into it. Citibank had been a car until I got in, then it became my car. My Citibank car.
But I was able to do a form of arbitrage by not attaching myself to the car itself. Once detached I could predict the car without trying to drive it or feel any attachment to its movements.
One could apply a methodology in two ways----choose a car and ride it long term. Or choose a methodology and ride in the car to a set point on a certain kind of stock or on stock that fit another variable.
The questions answer is possible but it requires narrowing ones perspective and detachment. Similar I expect to what banks, investment firms, hedge funds do. They focus on several car lanes and only shift as they see the road changing or an incursion from other lanes/cards in the future.
I also practiced a lot with mock massive market programs like Hsx.com which simulates the scope of the market but allows the detachment to your initial $2 million investment. I've turned that into almost $700 million but I have to say that in 17 years I only pay attention to it an average of 2 months a year. It should be billions but there are investment limits to avoid investor swaying, that limitation thereby creates a cascade limit on the fluidity of all of those millions, $60 million about the max on a movie one can invest at $600 a share. But the market itself in its artificial bubble is effected by investors getting in because more investors demand more buyable shares in the market.
Perhaps then that's the answer to a whole predictive model, there aren't enough humans in the game for a better perspective on movement. Think of it like fruit. From an aerial perspective we can see the amount of apple cars, orange cars and kumquat cars to see the market differences. But using the same analogy, there are too many individuals as cars to see clear trends and not get lost in fringe reactions.
We need more people in cars on the road, perhaps even smaller pools of shares being our fringe group as short term buyers, under 20, 10, then 1-5 years.
Following this, what if US social security were transferred to brokerage accounts? Some sort of investment training centers set up at libraries and then you're responsible for the outcomes? Say we started this for everyone under 21, drawing in 40-60 million youth based on their FICA contributions as they start working. Would that infusion of investors be enough to create an algorithm?
Could we then helicopter see the movement of the market in a predictive way?
No comments:
Post a Comment