Kimmel explores new models for studying interest rates

Formerly a Princeton economics professor and a researcher at the university’s Bendheim Center for Finance, Bob Kimmel left behind an institution that boasts influential economists tapped for high ranking federal posts and Nobel Prize winning scholars.

Yet Kimmel, an expert in econometrics and financial analysis, feels at home with his Fisher colleagues in the finance department. He is finding that a business school is a better setting for his research.

Kimmel’s research focuses on mathematic models used by Wall Street and academic researchers to track and test interest rate behavior. “The modern school of thought on interest rate modeling began in the 1970s and 80s,” he said. “In the 1990s, the affine yield models came into prominence and have been widely used.” However, many now argue that these models have systematic problems in capturing real world interest rate behavior. Kimmel’s work is dedicated to finding better alternatives.

The finance professor shared with Knowledge Link the latest developments in his studies.

KL: What is the focus of your research work?

Kimmel: “The main thrust of my work these days is developing new types of models for the behavior of interest rates. The affine yield model is one particular family of models for interest rates which are currently being used.

But no one really believes that these are the true models for the behavior of interest rates. People use them anyway because it is all we have to work with. Other types of models are too hard to work with to do the work of determining the price of derivatives or securities. My work is trying to make other families of models more usable.

KL: Why are affine yield models falling out of favor?

Kimmel: “There is a huge amount of research both in academic circles also in industry. This is an area where there is a lot of synergy between the two worlds. Evidence began to surface that they can’t get everything about the behavior of interest rates correctly. The affine models were not getting the volatility, particularly the volatility of long-term yields. In some instances, they tend not to get the average or the forecasts right. They haven’t been horribly inaccurate, but there are some systematic deviations from fact.”

KL: Which models do you think are more suited for interest rates?

Kimmel: “People have started to explore alternative families of models, but only in special cases. No one is clear yet on what are the right alternatives.

I don’t have one theory. I have several. This paper, “Complex Times: Asset Pricing and Conditional Moments Under Non-Affine Diffusions” is a unifying paper that looks at all the models explored in special cases and others that haven’t been explored. The purpose of my research is to make it possible for other researchers who are more empirically inclined than I am to be able to answer that question. They don’t have the tools to do it. So I am giving them the tools.”

KL: How is your work being applied? Are other researchers using your tools?

Kimmel: “A group of researchers were trying to look at the behavior of the prices of corporate bonds that have call options. They found that the technique I developed helped them solve a problem they were facing for a paper they were writing for a research article. With any such luck, there will be more application in the future.”

KL: What made you decide to leave Princeton to come to Ohio State?

Kimmel: “For the last few years, I’d been in an economics department. (Princeton does not have a finance department or a business school.) I feel like I fit in better at a business school. This is a high quality business school and I knew a lot of people here. The finance department is highly ranked for its research and gets things published in top journals.”