What do you think?
Rate this book
Use Data Analytics to Build Your Own Stock Fund and Beat the S&P 500
The short book is an introduction to the use of data analytics for identifying undervalued stocks which can be combined into customized investment funds. The methodology relies on price histories and business fundamentals made public through mandated financial filings.
Of course, the stocks that were undervalued as this book was written won't necessarily remain that way for long. The idea is to apply the methodology you'll see here to build your own funds. You can use the same Python code within Jupyter Lab notebooks (available for free from the associated GitLab repository) to apply the system. Not a Python fan? It'll work just as well no matter what data tools you prefer.
The systemGiven the resources of the modern internet, you can quickly and easily access detailed fundamentals data describing all 3,200 stocks traded on the major exchanges. With that, using any of the many mainstream analytics tools available, you can find just the companies that rate the highest among their peers using specific metrics.
In other words, let's say you suspect that the share prices of companies with extremely low "Liabilities to Equity Ratios" are likely to rise over the next few years, then you might select just the 20 companies rated highest for that metric. You can just as easily filter for a combination of multiple metrics and go with the 20 companies who best match all of those standards. You could also further diversify your portfolio by choosing companies from a range of industry categories.
When you're done, you'll have built yourself a sophisticated and highly customized and (hopefully) high-performing fund portfolio.
How do you know it will work?Well that's the beauty of having lots of historical data, as it makes it possible to test our assumptions.
Here's how that'll go. As you'll see in the chapter "Testing Your Model", you can use historical fundamentals data from previous years, create hypothetical stock portfolios based on any metrics you like, and then pull their actual market prices. The portfolio profiles that have consistently performed well in the real world in the past, are likely (although, of course, not guaranteed) to do so in the future.
How should we define "performed well"? Well I'd say that, over the past century, the blue ribbon marker for stock market success has been the S&P 500 index. If our funds can outperform the S&P, then we're in pretty good shape.
Of course, the stocks that were undervalued as this book was written won't necessarily remain that way for long. The idea is to apply the methodology you'll see here to build your own funds. You can use the same Python code within Jupyter Lab notebooks (available for free from the associated GitLab repository) to apply the system. Not a Python fan? It'll work just as well no matter what data tools you prefer.
The systemGiven the resources of the modern internet, you can quickly and easily access detailed fundamentals data describing all 3,200 stocks traded on the major exchanges. With that, using any of the many mainstream analytics tools available, you can find just the companies that rate the highest among their peers using specific metrics.
In other words, let's say you suspect that the share prices of companies with extremely low "Liabilities to Equity Ratios" are likely to rise over the next few years, then you might select just the 20 companies rated highest for that metric. You can just as easily filter for a combination of multiple metrics and go with the 20 companies who best match all of those standards. You could also further diversify your portfolio by choosing companies from a range of industry categories.
When you're done, you'll have built yourself a sophisticated and highly customized and (hopefully) high-performing fund portfolio.
How do you know it will work?Well that's the beauty of having lots of historical data, as it makes it possible to test our assumptions.
Here's how that'll go. As you'll see in the chapter "Testing Your Model", you can use historical fundamentals data from previous years, create hypothetical stock portfolios based on any metrics you like, and then pull their actual market prices. The portfolio profiles that have consistently performed well in the real world in the past, are likely (although, of course, not guaranteed) to do so in the future.
How should we define "performed well"? Well I'd say that, over the past century, the blue ribbon marker for stock market success has been the S&P 500 index. If our funds can outperform the S&P, then we're in pretty good shape.
44 pages, Kindle Edition
Published March 27, 2022
1 person want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
No one has reviewed this book yet.