Alexander Aue
University of California-Davis, CA·Mathematics & Statistics
Some data come as whole curves rather than single numbers. Examples include the yield curve, which economists read for signals about growth, inflation and interest rates, and the curve of a stock's returns over a trading day.
- Fields
Mathematics & Statistics
- Topics
functional time series
functional data analysis
nonlinear time series
- Grants
- 1 known grant, through Jun 2028
What they work on
- Derive new probability results describing the structure of nonlinear, possibly heavy-tailed functional time series, where each observation is an entire curve.
- Introduce a fully general functional GARCH model, a model of changing volatility, for curve-valued data such as cumulative intraday returns of financial assets.
- Develop a new functional random coefficient autoregressive model, together with inference procedures for analyzing it statistically.
- Build forecasting algorithms that are both theoretically justified and validated on data, for example to predict volatility curves from tick-by-tick financial transaction data.
More about the research
When such curves are recorded day after day, they form a functional time series. Most existing tools assume these series behave linearly and follow a bell-shaped (Gaussian) distribution. Real financial data often break both assumptions: their behavior is nonlinear, and they are heavy-tailed, meaning extreme values are common.
Why it matters
Better forecasts of curve-shaped data could help finance practitioners predict the intraday volatility of assets. They could also help economists interpret yield curves as signs of future growth, inflation and interest rates. The methods could be applied in many other fields that rely on data-analysis tools.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Benjamin Boniece
Drexel University, PA·Mathematics & Statistics
Develop methods to detect shocks that affect many series in high-dimensional, high-frequency data, possibly asynchronously, using time-localized principal component analysis in factor-structured models.
Ting Zhang
University of Georgia, GA·Mathematics & Statistics
Develop change-point tests that stay powerful when the post-change signal varies unpredictably, using a causal representation framework and Gaussian approximation to set statistically sound alarm thresholds.
Jyotishka Datta
Virginia Polytechnic Institute and State University, VA·Mathematics & Statistics·Early-career
Develop Bayesian independent component analysis with heavy-tailed priors and novel latent space representations, unifying feature extraction methods and controlling higher-order moments when separating hidden sources from mixtures.
The school
University of California-Davis: 49 labs on PI Pages, $1.0B a year on research, 647 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.