Thread by @joelalcedo1

3yrs ago, I uploaded econ/finance R coding tutorials but never kept up with it

I recently discovered one reached 5,000 views - surprising to me

Inspired these could help others, I'm giving it another go, will try to upload 1x/month

Here's a new vid!👇

In this tutorial, you'll learn how to:

1. scrape historical #cryptocurrency and #stock price data

2. calculate daily returns

3. identify uncorrelated assets using a hierarchical clustering algorithm

4. backtest an uncorrelated portfolio of stocks against a correlated one

Why does this matter?

Stocks, currencies, bonds - basically any asset - can go up/down for the same reasons (eg. same macro factors, policies, a pandemic)

An uncorrelated/diversified portfolio is an important tool to improving the likelihood of better risk-adjusted returns

An analogy

Strawberry jam & jelly are different foods made of the same fruit

If you only eat jam and jelly, you'd likely end up with strange health issues

Stomach space is finite - efficiently allocating a broad range of nutrients mitigates the risk of health issues 😀

The very basic implementation in the tutorial illustrates diversified portfolios tend to have favorable sharpe ratios and annual returns compared to a highly correlated portfolio.