Quantitative Investing

Quantitative Investing

At the time of this writing, in July 2022, we are experiencing unprecedented economic conditions. Most of the assets we purchased as investments are losing their values to make us feel poorer, and inflation keeps rising to make us actually poor. I often get DMs on Discord asking me, "I am losing so much money. Should I sell and buy back later?" or "The price of everything went down so much. Is now the right time to buy?" The answer, of course, is "don't try to time the market." However, I don't want to stop there but elaborate more on this. This touches upon such a fundamental shift in one's perspective on how one understands the world and investing in general.

Model vs. Reality

Beginner investors all want a clear, simple answer from me to follow. Basically, when to buy what. The fact that people ask such a question tells me we tend to grossly underestimate what is actually going on in reality. In reality, the prices of assets are presented to us in deceptively simple line charts, but the reasons why they move the way they do are infinitely complex and chaotic. To become a good investor, you must stop being incredibly naive to think that you should try to somehow time the market.

Throughout our lives, we are taught to define problems, objectives and solutions not from reality itself but from a model of reality. A high school physics exam question asks you to calculate the force of gravity applied to the ball being dropped by observing its fall velocity. It conveniently omits what material the ball is made of and how the wind blowing on it can affect the speed of its fall, which could ultimately result in the incorrect answer. Most students memorise the formula and try to get the "right" answer, but rarely do they think about the ridiculous number of assumptions made in the question itself, which is a grossly simplified representation of reality. Interpreting and analysing nature and reality through a model is a perfectly valid approach because, after all, nobody has the time and resources to consider every single variable in the real world for a given situation.

What investing decisions really are

In investing, we are simplifying the reality of a company into a modelled representation of what they are:

  • Income statements
  • Balance sheets
  • P/E ratio
  • Debt-to-Equity ratio
  • Goldman Sachs's official rating of the stock
  • etc.

We try to make the "right" decision from our model and apply them directly in reality by purchasing the stocks. However, we forget that no matter how perfect we think our models are,  there are always hidden variables we are missing:

  • An unforeseen pandemic that can cause a national lockdown and cause the entire business operation to a halt
  • CEO is secretly cooking the accounting books, which gets caught
  • etc.

It is virtually impossible to know and consider every single variable that could affect the future performance of a single company. This is also the reason why it is practically impossible to manually pick individual stocks and expect to outperform the index year after year for more than 10 years. If you are modelling and analysing individual companies, the gap between the model and reality is HUGE. However, suppose you are investing in diversified ETFs across many companies. In that case, all the real-world noise each company brings gets averaged out, and the gap between the model and reality becomes much smaller.

Ultimately, no matter what analysis you do, you are still making decisions based on the modelled reality. A great way to find out you're mistaken and immediately improve on your model is to use various statistical and quantitative metrics because they are much more precise and detailed than anything you can write in natural languages like plain English.

← Guides