Knowing Institutional Datasets Makes You Extremely Valuable
Introduction
If you’ve been in a megashop you would know that the amount of datasets (vendors, products) is immense. If you are in a small shop, you probably would not come close to even understanding the vast amount of data that goes into a systematic investment process. This is the MOST comprehensive collection of real data categories that are actually used in systematic strategies.
Most systematic managers start with price and fundamentals. Some add sentiment. A few experiment with satellite imagery or credit card data. The managers who win are not necessarily the ones with the best model. They are the ones who understand the full range of data available and deliberately choose which categories to exploit.
The systematic feature universe spans at least 20 distinct data categories - each with different signal characteristics, different decay profiles, and vastly different implementation costs. If you are only using two or three of these categories, you are leaving diversification on the table. And in a world where the signal-to-noise ratio is about 1%, diversification across data types is one of the few reliable sources of edge.