Thread by @TechMemeKing

In honor of #Bitcoin2023 I examined the log log model of Bitcoin’s historical price by using ChatGPTs Code Interpreter. 👇🏼👇🏼

Power law approach and model originally by @hcburger1 🙏🏼

Using daily price data from 2010 through 4/26/2023 I asked code interpreter to fit the total data set using @hcburger1‘s methodology.

It projected the model out through 2030. 👇🏼

The model fit through 4/26/2023 produced an R2 value of 0.953.

A very good fit. It correctly reminded me that a high R2 value does not guarantee future model predictions.

Here are some of its predictions. (Not financial advice):

These can be interpreted as fair value price predictions as they are based on the best fit. It doesn’t attempt to predict daily prices accurately.

I then attempted to recreate @hcburger1’s model by only fitting the model through 12/30/2019.

Amazingly, the model through 12/30/2019 had an R2 of .953 (almost exactly the same as fitting all data through current).

Model fit through 12/30/2019 in log log scale.

The model has predicted quite well since he originally fit it. 🔥

Just for fun I asked Code Interpreter to add prediction intervals for 1 and 2 standard deviations.

It decided to use the standard deviation of the residuals from the fit of the model.

I’m amazed at how easily you can analyze data with Code Interpreter. I didn’t write a single line of code.

Many thanks to @hcburger1 for the inspiration. His Power law Bitcoin model is the best out there (imo).

If you enjoyed this, an RT or a follow would be much appreciated!