Economic Forecast Algorithm: Future or Now?Posted: April 04, 2012
The future is something we all think about. It captures our imagination through both our fear and excitement. But what if we could predict it? What if we could avoid economic failure and save peoples’ livelihoods?
Over the last few months I put together an in-depth analysis on how taxes (specifically capital gains) played a role in the 2008 stock market crash and the subsequent great recession. During my time researching the economy I was frequently bothered by one particular thought:
Our knowledge of economics, abundance of historical statistics, and availability of computer technology is far too advanced to be making such grand mistakes.
The answer to this problem is simple: create an economic forecasting algorithm.
So What’s an Algorithm?
The technical definition of an algorithm “is a specific set of instructions for carrying out a procedure or solving a problem”. (Source)
Pretty vague, but you get the idea. It’s an equation created to take a large amount of inputted data and output it as usable information.
Is it possible to create an algorithm so advanced it could accurately predict economic outcomes? To answer that question let’s look at three complex algorithms currently in use.
I work in the field of search engine optimization. This means everyday I analyze and reverse engineer the many algorithms Google and other search engines use to populate their organic results. I deeply understand the complexity of algorithms and what they’re capable of computing.
Each year, Google alone changes its algorithm 500-600 times. (Source) Search engines also implement a web of different algorithms simultaneously including local specific, browser specific, and user specific variations.
There are even examples of how search engine algorithms have similarities to what an economic algorithm would look like. In one example, link juice and/or page rank flow through different websites in a rough but similar fashion to how currency flows through different countries.
Not too long ago I was having lunch with my friend Lynn and her dad. He was talking about the work he was doing for his company InstinctCode. InstinctCode is a software company that creates thermodynamic algorithms.
On one project he had created software that uses air and water thermodynamics in order to better predict weather patterns across the globe. The software has the ability to greatly improve many weather mapping systems.
Sounds impressive right? Well this is the software his company was providing free and open sourced from their website. The projects currently underway are much more complex.
Bruce Bueno de Mesquita:
Bruce Bueno de Mesquita is a political scientist, professor at New York University, and senior fellow at Stanford University’s Hoover Institution. He founded a company, Mesquita & Roundell, that specializes in making political and foreign-policy forecasts. (Source)
Bruce used game theory to create a computer algorithm that predicts future events. In essence he takes the current political environment to predict the future actions of each country.
Just how accurate are Bruce’s predictions? “Two independent evaluations, one by academics and one by the U.S. Central Intelligence Agency, have both shown that about 90 percent of his predictions have been accurate.” (Source)
This is a great example of how events influenced by psychology and sociology can still be quantitated and predicted using computer algorithms.
Economic Forecasting Algorithm
Things like tax rates, monetary rates, government spending, and national debt are all variables that can be entered into an equation. They’re all dependent on each other and fit into a global economic algorithm. All these variables can be used to calculate how changing one or another will affect the way money moves through an economy or group of economies.
The question is no longer ‘can it be done?’, but has it already been created?
Brian Rogel isn’t a household name but his analysis of wealth inequality could become household reading.