Thought Leaders in Financial Technology: Infima Founder and Chief Scientist Kay Giesecke (Part 2)

Sramana Mitra: What are some of the nuggets that you’ve learned?

Kay Giesecke: We learned that the behavioral patterns are very complex. Let’s just focus on homeowners. There are the lenders. They look at applications for mortgage loans. They need to decide if this person is creditworthy enough for a home mortgage that’s backed by a specific home that they’d like to purchase. They have to assess the chance that this person is going to be able to repay the mortgage on time over the term of the mortgage.

They need to factor in all the possible macroeconomic scenarios that may unfold over this time period. It’s a difficult projection problem. It really is a complex pattern that emerges. No two people are the same. Things depend on their individual circumstances and the macroeconomic scenario that they are in. It’s these conditions and how they interact, that’s the major insight that we’ve obtained.

Sramana Mitra: Your algorithm must have clustered consumer behavior. There are macroeconomic conditions that vary. That’s one set of variables but your algorithm must have done some amount of clustering to be able to predict.

Kay Giesecke: These are classification problems. Someone can be current or 30 days behind. They can be in forbearance. We’re tracking all of these states that someone can be in and try to project how someone is moving between any of these states over future time periods. That is strongly dependent on what the macroeconomy does. The macroeconomic effects different people in different ways. That’s where the complexity comes in.

Sramana Mitra: Not only that. We have seen a once-in-a-century situation over the last three years, which we can’t model.

Kay Giesecke: Absolutely. This is where we’ve made a lot of progress. We’re able to crunch a lot more data, get all the complexity built into those models, and make more accurate projections even in these very volatile time periods.

Sramana Mitra: What are the different buckets in which your models operate? One is mortgage finance. What are the other buckets?

Kay Giesecke: Currently, the focus is on mortgages, but we have done work on the academic and corporate side. If you’re thinking about the big public companies but also the small businesses like restaurants and barbershops, the technology has been proven to work very well in those verticals.

Sramana Mitra: But the company is focused on the mortgage finance application?

Kay Giesecke: It’s focused on the mortgage piece. There are other pieces including companies big and small as well as municipal agencies like the city of Palo Alto raising bonds to finance a project. How is Palo Alto going to be able to repay that money?

Sramana Mitra: Your company focuses not only on consumer mortgages but also corporate mortgages and property purchases.

Kay Giesecke: The technology has been built, but, at this point, the company focuses on mortgage exclusively with a plan to expand to these other areas. There’s a clear path to expand into other verticals, but we haven’t done so yet.

This segment is part 2 in the series : Thought Leaders in Financial Technology: Infima Founder and Chief Scientist Kay Giesecke
1 2 3 4 5