This is the Cato Daily Podcast for Saturday, November 30th, 2019.

I'm Caleb Brown.

When the Federal Reserve tells us what they're doing, the language they use often obscures exactly what they're

doing.

Charles Calamyrus teaches finance and economics at the Columbia Business School at the Cato

Institute's Monetary Conference held earlier this month. We talked about what natural

language processing could mean or how central banks communicate.

Most people don't really understand what the

incentive to try to dig into what it is they do not lend people

the incentive to try to dig into what it is they do. So tell me what is natural language processing and how

ought it to influence Fed communications.

So all natural language processing means is that we're taking word flow,

whether it's fed speeches or press releases or minutes,

or news articles that appear in newspapers or whatever word

flow and we're analyzing it which means turning that word flow into something quantitative.

So we're looking to see can we learn things from all these different word flows

that can help us understand something in a disciplined way, in a quantitative way.

Can we take the sentences that you just spoke to me and in some sense boil them down to some set of numbers about something.

So that sounds kind of strange to most people, right?

But actually that's been a revolution over the past 10-15 years in social science and it's going to

revolutionize even more.

So all of social science, without any hesitation predict will be completely revolutionized by this new set of

...