What Should I Name This Index?

For our first post, I want to at least figure out the name of the index I'll be designing.

The Berkshire Buffet Index

We left the last post on a small cliffhanger. Here’s where we’re actually going with it. I’m calling the index the Berkshire Buffet Index — a buffet of sorts where the best items are picked with a value-focused mindset.

In our upcoming posts, we’ll break down how the index is built, what qualifies it to borrow a name from the great Warren Buffet(t), and how it stacks up against popular growth indices like the S&P 500. We won’t be chasing every trade — instead, we’ll ask, for each security, if it passes our smell test for a Buffett–styled Berkshire Buffet Index?

Ingredients For Our Berskhire Buffet

For today’s post, we’ll discuss a bit about how we’d like to pick our securities. Initially, the thought process was to measure securities based on unidentified worth, but that's extremely difficult, especially in a world where information is so readily available. So, what if you could combine sentiment analysis and web data scraping to assess companies that hold the most “perceived value by communities online,” and then rank these securities based on the general sentiments held about them? Thinking about this poses a lot of questions. First, how much data will we need? If we're evaluating the entire universe of securities, there are roughly 8,000 securities from various markets. Will we need to measure every single one of these—to gather thousands and thousands of lines of data to make the “best” assessment of market sentiment about that particular stock? But what if we had an intermediary who could process these large amounts of data? Would that allow us to measure sentiments across multiple social and internet-based platforms? Well, actually, one could simply ask: why not ask the AI bots of these systems what those sentiments are? You could ask whichever one you choose to provide a random sample of posts, per se, for each security. But it still begs the question of the data processing needed to assess all 8,000 securities. In that case, segmentation might be our answer. We'll design metrics to reduce this number to a manageable amount to allow us to scrape our data. So, for tomorrow—or today—what will we be doing? Designing those metrics by which we'd like to evaluate the sentiments of our stock.

Buffet Metrics