Aurore Delaigle is a Professor and ARC Future Fellow at The University of Melbourne. Before moving to Melbourne in 2007, she held positions at the University of California at Davis and at San Diego, and at the University of Bristol.

Aurore’s main research interests include nonparametric estimation, measurement errors, deconvolution problems and functional data analysis. She is particularly interested in developing new methods for indirectly and imperfectly observed data, especially if they have some seemingly unappealing form (e.g. incomplete, irregular, nonstationary, too big to handle with usual methods).

In 2013, Aurore was elected a Fellow of the IMS and was awarded the Moran Medal from the Australian Academy of Science. She is an associate editor for some of the most prestigious international statistics journals. She was the Executive Secretary of the IMS from 2011–17, and was the Associate Program Chair of the 2017 Joint Statistical Meetings.

This Medallion Lecture will be given at the Joint Statistical Meetings in Toronto, August 5–10, 2023.

Estimation of the density of a long-term trend from repeated semi-continuous data, with applications to episodically consumed food

In this talk we consider semiparametric and nonparametric estimation of the density of the long-term trend of a semi-continuous variable observed repeatedly over time. Variables of this type arise when measuring the intensity of an intermittent phenomenon, such as the intake of an episodically consumed nutrient measured through repeated 24 hour recalls or the concentration of an intermittent toxic substance: on a day where the phenomenon is absent, the measurement is equal to zero; otherwise, it takes a positive value. Unlike daily consumed nutrients which are typically represented by a classical measurement error model, data with clumping at zero are usually represented by a two-part model describing the zeros and the non-zeros separately, and connected through latent variables. Several variants of the model and methods have been proposed under parametric assumptions. We study more flexible non and semiparametric approaches to this problem, which can be viewed as one with a combination of measurement errors and excess zeros.

JSM, the Joint Statistical Meetings, this year will be in Toronto, August 5–10. Check out the program at https://ww2.amstat.org/meetings/jsm/2023/. IMS highlights include two Wald lectures by Bin Yu; the Blackwell lecture (Ya’acov Ritov); the Wahba lecture (Wing-Hung Wong); four Medallion lectures, from Ingrid Van Keilegom, Runze Li, Yingying Fan, and Aurore Delaigle; and the IMS Presidential Address and Awards session on the Monday evening. This is alongside a packed program of talks, posters, roundtables, professional development courses and workshops, award ceremonies, and countless other meetings and activities. JSM will be held at the Metro Toronto Convention Centre (255 Front Street West, Toronto), with additional meetings and events at the Delta and InterContinental hotels. See you there!