Bayisian Probability 🎲

The past few weeks I’ve been reading a lot about statistics and probability. I read a book called Statistics for Dummies whose audience was clearly High Schoolers and I struggled through parts of it like I did many parts of Pre Calculus in High School and Elementary Statistics in College. I get the math but the symbols always throw me off. I can do quadratic equations but they aren’t my favorite.

Then I started to read Nate Silvers’ The Signal and the Noise. Much easier read for me, and more fascinating as it looked at real life statistics. It’s easy to be mislead by incomplete data and bias, and either underweighting or overweightingaaaa priors. But using Bayisian Probability can help one overcome some of the current bias by looking back over time. Things usually continue in their current fashion and extremes ultimately revert to the means.

It really is good to keep the past in mind when thinking the future. Past success can be built on. Risk exists but can be minimized and smoothed out over time. If yesterday was good, today is better than chances are good that tomorrow will be even better than today.

I’ve long been interested in risk management and seeing the world not through the fearful panicking nature of the news media, afraid of today and not considering long term structural risk and rewards. I’m a libertarian so I tend to prefer more risk to human life and well being in exchange for more human freedom. Caution can be warranted but it shouldn’t come at the cost of human freedom and dignity.

But my real interest in statistics today is more practical – I have powerful analytical tools like R Studio and access to a remarkable amount of open data that can better explain the world around me, if only I know and apply proper statistical analysis to it. I’m no Quant or statistician and have little interest in returning to formal education but I know there is much I can learn myself.