Consider the impact of higher policy rate on a day to day consumer decision like eating out. Possibilities are many: high, low and no impact. The impact can vary by region, demographics, ethnicity and income groups. Hospitality industry will be keen to understand the impact of change in policy rates or other macro factors and how they are impacting the daily lives of the consumers.

Macroeconomics is often described as ‘pseudo science’ and lacks granular insights. On the other hand, data science, ML and AI are too focused on nuanced problems restricting them from having a broader vision. 

But, the question is, can we combine these two to make insights more holistic and complete? 

AI/ML in economics 

Today, large tech firms have economists who use data to estimate key parameters that help them curate tech policy. The economists work very closely with data sciences, applied scientists and developers to help inform internal decisions by leveraging data. 

At the same time, ML with econometrics has also become an active research area in economics, thanks to the availability of large datasets, especially in microeconomic applications. However, applying ML to economics is easier said than done. To begin with, it requires finding relevant tasks. Moreover, the field is still nascent, and very little progress has been made to understand ML models’ properties when applied to predict macroeconomic outcomes. 

Econometrics applies statistical modelling that helps understand complex social and environmental issues. It is one of the emerging areas of applied economics.

The whole area of ‘econometrics’ is based on applying statistical models, including descriptive statistics and hypothesis testing, correlation and regression analysis, time series analysis and visualisation, predictive analytics and forecasting, panel data modelling (like OLS, fixed and random-effects models), etc. 

Economists’ job roles are similar to data scientists. For example, behavioural economists quantify consumer response and attitude to marketing campaigns, discounts, price reduction, etc. Similarly, industrial economists check the level of commodity production and pricing mechanism for given market demand.

How AI/ML can enrich macroeconomics

In macroeconomics, the predictions/forecasts made by economists always seem to be terribly wrong. This is where AI/ML comes into play. Predictive analytics can help overcome some of the challenges in economic forecasting. 

By leveraging AI techniques in behavioural economics, economists can accurately estimate the impact of human perceptions and behaviour. For example, JPMorgan used an algorithm to track the effects of President Trump’s tweets on financial markets. Also, central banks and financial institutions can be more effective by predicting recession hits, thereby mitigating the effects on the business. Economists can also bring changes in supply and demand and implement necessary changes to avoid economic turmoil.

Most importantly, based on the real-time data, economists can help companies curate technology policies to run businesses smoothly and avoid downturns.  

Why is AI-fication of macroeconomics important? 

While macroeconomics zooms out of the minute details and views the world as a large interconnected system, microeconomics focuses on the interaction between basic building blocks of the society and the result of the interaction. 

Some of the examples include – 

• How a local business decides to allocate their spends 
• How the government decides to spend the tax income
• The housing market of a particular location/city
• Production/manufacturing of a local business/product; etc. 

Simply put, microeconomics is related to decision-making with low-level effects, say, allocations of spending, location/city, etc. But, on the other hand, macroeconomics has a high-level, large-scale impact, say the business, government, housing market, production, etc.

Macroeconomic analysis can help data science to make insight generation more comprehensive. 

Here’s how 

• Spectrum of analysis: Capturing both macro and micro variables enables combining the impact of macro trends with the micro big data-driven insights. 
• Completeness of insights: Businesses can assess the impact of economic trend changes on very nuanced challenges that they face. They may be intrigued to understand gritty insights that shape the new age goods and services offering. 
• Societal benefit: The combined solution has greater potential to unlock value and impact our daily lives and our decisions towards broader societal change. 

A final thought

As the name suggests, macroeconomic analysis is macro in nature and explores the relationship between the macro variables and analyses the macro trends impacting the economies, public policies, and industries broadly. On the other hand, data science and AI targets industry or business-specific challenges, which are often restricted given certain micro assumptions. 

While macroeconomic analysis lacks granular insights, data science, ML and AI don’t have a broader vision. Thus, combining these two is a great idea to have a more holistic solution covering both the macro and micro sides of an industry or business pain points.

This article is written by a member of the AIM Leaders Council. AIM Leaders Council is an invitation-only forum of senior executives in the Data Science and Analytics industry. To check if you are eligible for a membership, please fill out the form here.