According to Nobel Prize-winning economist, Richard Thaler, there are a plethora of psychological fallacies out there that cause self-inflicted financial losses, with the following being particularly destructive when it comes to trading and investing…
The Endowment Effect:
A reluctance to sell things just because they belong to you.Recency Bias:
Thinking that whatever’s happened in the past will continue in the future.The Sunk Cost Fallacy:
A tendency to continue an endeavor because you’ve invested your money, effort, or time into it.Hindsight Bias:
Thinking that you predicted the unpredictable.
Knowing this, right now, wouldn’t it be great if we could eliminate almost all money-losing fallacies, especially when we’re trying to time both entries and exits; the part of the investment process where we become the most vulnerable? What if we could build a system that helps us to control our emotions, that helps us to sell when everyone’s buying and to buy when everyone’s selling, so we have the best chance of making money?
Of course, the perfect model, system, or formula does not exist and never will exist, but that doesn’t mean we can’t calculate probabilities. To find reliable, not perfect, entries and exits, I use my sentiment model, an algo that computes extremes in indicators to measure positioning and sentiment.
I was inspired to create this after reading a passage from the Man Who Solved the Market by Gregory Zuckerman:
“Investors generally sought an underlying economic rationale to explain and predict stock prices, or they used simple technical analysis, which involved employing graphs or other representations of past price movements to discover repeatable patterns. Simons and his colleagues were proposing a third approach, one that had similarities with technical trading but was much more sophisticated and reliant on tools of math and science. They were suggesting that one could deduce a range of “signals” capable of conveying and useful information about expected market moves.”
It was only then that I started to realize what was useful and what was noise. Put-call ratios and implied volatility, for example, told me a lot more about psychology and positioning than basic linear indicators such as SMAs and RSIs. So, using dynamic indicators, I built a model that tried to predict the upside and downside of an asset in percentage terms, for example, it might have discovered that Apple’s Stock had 10% upside and 90% downside based on factors X, Y, and Z.
That output, however, was unusable without translating it into a decimal range (for example 100.12 - 120.23), but by using an asset’s known market metrics such as realized volatility, my model could calculate a real-time minimum and maximum price that I could pair with my intuition.
You can find what my sentiment model says about all the major asset classes and sectors in the macro wall below. Notice how I have removed the minimum/maximum price ranges. That’s intentional because, of course, values change over time, and significant market moves will render the values in a newsletter useless. To counter this, slightly, I’ve added a visual range. I hope that helps.
Finally, if you’re interested in creating your own sentiment model, I have included the algos and indicators my model uses in the “stats for nerds” footnote below. It’s a steep learning curve but a worthwhile venture.
Concoda
Stats for nerds: Algorithms, indicators, and services used: ARIMA, VAR, Python, Anaconda, Scikit Learn, fBM, Volatility Studies, Volume Studies, Momentum Studies, Currency Correlations, Long Term Dependence, TWS API, various vol services
If you act on anything provided in this newsletter, you agree to the terms in this disclaimer. Everything in this newsletter is for educational and entertainment purposes only and NOT investment advice. Nothing in this newsletter is an offer to sell or to buy any security. The author is not responsible for any financial loss you may incur by acting on any information provided in this newsletter. Before making any investment decisions, talk to a financial advisor.
Nice.