A Model for Initial Jobless Claims

The Dept. of Labor publishes data on initial jobless claims weekly, and today much of the talk in the financial media was about this on the grounds that, in the era of Coronavirus, slow metrics like Non-Farm Payrolls are not useful in understanding how the economy is responding now.

Although we run consumer surveys that ask about respondents employment situation, we’ve never asked directly “did you claim for unemployment insurance this week” and so any data we have access to is bound to be a little indirect. We do ask about “job seeker” status, and so differencing that datum, and multiplying it by the eligible population size, should be indicative of “new claims.”

The problem is that I don’t know what the eligible populate size is. However, if I assume that relationship between the initial claims number and the data we have exhibits some kind of local temporal stability then I can use the sort of predictor-corrector machine learning model we’ve previously applied to Non-Farm Payrolls to figure out what that is.

So, cutting to the chase, this is what we get:

I’m reasonably pleased with this, so will be adding it to the data collection that is available for subscription on AWS Data Exchange. I’ll post the link when that is done.

Alternatively, you can order the paperback directly from me via our website.

You can directly support my writing on Medium by subscribing through this affiliate link.

Related posts:

Because inspiration comes through the words and works of well-known poets and authors.. “This Poem is My Daily Dose of Motivation” is published by A Rustic Mind (Manali Desai) in Cutting Chai.

His convictions are so strong that he plans to merge the messaging services of Facebook, WhatsApp and Instagram into one encrypted system, meaning only the people who send and receive messages can…

The advantages of making slow reading a normal habit are numerous, reducing tension and improving your capacity to concentrate. The advantages of reading book regularly also have improvement in…