This blog post is about a book titled “Predictive Edge” written by Alejandro Lopez-Lira

I picked up this book after listening to someone referring to the author in a talk. This book has taken me forever to read mainly because it contains 230 pages of just text, more text and more text. There is not a single visual, math equation or any graph and it makes the book so very boring to read. I read a few chapters many months ago but then put it aside as I found the content in the initial chapters very underwhelming. For a book that advertises itself as providing a predictive edge, one really has to skip the first 134 pages to get a glimpse of what the author has to say. Let’s say you do decide to skip the first 134 pages to get to the strategy, you will be severely disappointed as the strategy is mentioned at a 10,000 ft view and that too, in words that makes reading this book painful. It looks like the strategy is written by an intern who is fresh out of college and has picked up some cursory knowledge of finance. May be it is better to read the paper written by the same author

I can’t believe that I will be paying late charges to the library as it does not deserve the time at all. The only reason I kept the book beyond the library due date is : I somehow wanted to read cover to cover and see if I can at least get something about the book. Sadly the book is total waste of time and money. For any reader with even reasonable financial markets experience and trading experience, there is nothing to gain from this book. Sometimes I wonder whether the content itself is ChatGPT generated using a few smart prompts. Every strategy is using sentiment signals and somehow author seems to be saying that LLMs have revolutionized sentiment analysis. Tons of data vendors have been providing sentiment data and the relevant unstructured data for decades.

Skip this book and you will not miss a thing. The book DOES NOT live up to the hype associated with the paper that is mentioned in the book. May be you are better off downloading the paper and see if the content makes sense