Is Correlation Useful?
We want to know how independent the returns of assets in our portfolio are. Most people use correlation to measure this. Should they?
Read more
Probability and Statistics
This Stat Pisses Me Off
I always see people posting something along the lines of “the market has gone up in a 20-year period 90% of the time”. This drives me insane.
Read more
Statistics Is Useless Part II: The Consequences
Statistics is hard. It’s hard to tell what generating function we are dealing with. And this function can change. We must use what we know via deduction.
Read more
Statistics Is Useless Part I: The Problem
Distributions come from one of four types of generating functions. The problem it’s hard to tell what the generator is. This causes all types of problems.
Read more
Why Things Happen
How do we know why things happen? Statistics can’t explain. Neither can mathematics. So how do we determine causality?
Read more
How to Lie With Data
Data lies sometimes. How? Sometimes the inputs are faulty. Sometimes the comparisons are unfair. Visualisations can be misleading.
Read more
A Quick Post on Deduction
Deduction, as opposed to induction, is the act of arriving at conclusion bottom-up, by applying logic to axioms. Here we introduce the technique.
Read more
Induction: Hume vs. Keynes
Keynes argued against Hume’s perspective on induction. But they were actually talking about slightly different things.
Read more
Owning vs. Renting: A Definitive Answer
Is it a better financial decision to buy a property or rent? Buying, probably – but it depends on the situation. You have to run the numbers. And even then…
Read more
Gareth’s Big Mistake: Distributions vs. Averages
Looking at moments restricts some information. Focusing on single-point estimators, rather than the while sample, can lead you to mistaken conclusions.
Read more