Posted by David Kretch on 7/21/15 11:30 AM
Read more about me: Biography
* This is the first post in our new blog series: Predictive Analytics in Enforcement.*
People and organizations often need to make decisions that depend on what has happened, what is happening now, and even what will happen. We guess as best we can, but decisions must be made every day using less-than-complete information. It’s tough to make predictions, especially about the future—prescient words from little-known 20th century American sports figure Yogi Berra. Predictive analytics describes the practices and techniques used to make the most well-informed guesses possible given the information available.
Nearly all aspects of society rely on predictions. Governments need to know the costs and benefits of public policy, how much revenue they can expect, who is violating regulations or breaking laws, and the overall direction of the economy. Businesses need to know what their customers want, how much product they need to order or manufacture, and where to put it all. It is often prohibitively expensive, impractical, or (in some cases) impossible to measure all of these outcomes—so we predict them instead.
Predictive analytics uses tools from a variety of fields: probability and inference from statistics, machine learning from computer science, and econometric modeling from economics. This process is used every day by many different organizations:
Google predicts what web pages you want to see and ads you’re likely to be interested in, based on your search queries and how web pages link to each other.
Netflix predicts what shows and movies you’d like and Amazon predicts products you’d be interested in, based on what you and others with similar tastes have liked previously.
Predictive analytics is also used by governments in many applications, from making operations more efficient to aiding enforcement:
The Internal Revenue Service detects tax fraud, identity theft, and other kinds of non-compliance.
Here at Summit, our team helps implement predictive analytics capabilities for Federal agencies including Treasury’s Debt Management Services, the Mine Safety and Health Administration, and the Department of Labor’s Employee Benefits Security Administration.
As we can see, the tools of predictive analytics are useful in many different applications. As we continue this blog series, we’ll discuss in more detail how they work and how they can be used in government, particularly in enforcement of laws and regulations.
For more information about predictive analytics, check out Summit Principal Albert Lee's recent publication in the AGA's Journal of Government Financial Management: "Predictive Analytics: the New Tool to Combat Fraud, Waste and Abuse"