Archives of “economic model” tag

rss

What U Can Learn From Occam’s Razor About TRADING

  
  Education
#  #  #  #  #  #  #  #  #  #  #  #

I have seen too many traders that randomly add condition after condition to their trading strategy, hoping that it will increase their hitrate. What they are trying to do is to add assumption after assumption to their hypothesis, until their hypothesis (“price will move in to this or that direction for this or that amount”) is hopefully correct more often than not.

Going this way usually ends in paralysis through analysis or in total chaos because there are so many conditions when entering a trade that it is impossible for a human brain to follow the system, thus inducing mistakes.

 Enter: Occam’s Razor

Occam was one of those scholastic philosophers, living around 1300 A.D. He developed a principle called Occam’s Razor which states that “among competing hypotheses that predict equally well, the one with the fewest assumptions should be selected. Other, more complicated solutions may ultimately prove to provide better predictions, but—in the absence of differences in predictive ability—the fewer assumptions that are made, the better.” But Occam was not the first, even Aristotle, who was living a mere 2000 years ago, theorized about this concept.

Now the thing is that of course when you add or remove certain conditions to your trading strategy, the predictive ability of your strategy will vary thus rendering Occam’s Razor seemingly invalid. Seemingly. Because trading is a game of incomplete information, we can never exactly know the predictive ability of a model.

Even after testing thoroughly, we will always only get an estimate (our winrate). Because of this fact you should base your trading strategy on as few assumptions as possible. (more…)

Fascinating Insights From Nobel Prize-Winner Robert Shiller

  
Analysis, Education
#  #  #  #  #  #  #  #  #  #  #  #

On why so many experts missed the 2008 financial crisis: “Experts have always missed big events like this. If you look at the record of statistical forecasting models, they tend to get to the recession when it’s starting to come. A casual observer might start to worry about it. Forecasting it years out, they don’t get; in particular, if you look at the Great Depression of the 1930s, nobody forecasted that. Zero. Nobody. Now there were, of course, some guys who were saying the stock market is overpriced and it would come down, but if you look at what they said, did that mean a depression is coming? A decade-long depression? That was never said.”

On short-term thinking: “I think that there’s too much faith in analysis of short-term data. You see some pattern, and you can do a statistical test and prove that will prove that it is significant or passes the smell test to a statistician. But the problem is, the world is always changing. It’s not a stable thing. The underlying human parameters may be stable, but you can see that there is institutional and cultural evolution, and it’s not something that you can quantify.”

In Markets, Here Is No Why

  
  Education
#  #  #

Markets are and will always be inefficient because they are driven by narrative-derived expectations.  Certitude behind the reason why a market or instrument’s price prints where it does at any time isn’t possible, so what is more or less probable is deemed sufficient.  Reliable probabilistic assignment of causes has difficult methodological problems and is usually much too impractical to provide a sound procedural basis for active market decisions.  Crude narrative formation or adoption from mass outlets is typically substituted; and disappointed expectations often the result.  One path of remedy is creation of tractable yet rigorous models scoring pertinent fundamental, technical, sentimental, quantitative and news-based variables; while another is renunciation of narrative-based analyses and explanations: in effect, a renunciation of the analytical pursuit of why.

Go to top