This site will look much better in a browser that supports web standards, but it is accessible to any browser or Internet device.

Welcome to David van Dyk's web site photo: dvd


Advances in astrophysics and cosmology are driven by large and complex data sets, which can only be analyzed, interpreted, and understood by using refined statistical methods. In recent years, technological advances have dramatically increased the quality and quantity of data available to astronomers. Newly launched or soon-to-be launched space-based telescopes are tailored to challenges associated with specific scientific goals. These instruments provide massive new surveys resulting in new catalogues containing terabytes of data, high-resolution spectrography and imaging across the electromagnetic spectrum, and incredibly detailed movies of dynamic and explosive processes in the solar atmosphere. The range of new instruments is helping scientists make impressive strides in our understanding of the physical universe, but at the same time it is generating enormous data-analytic and data-mining challenges. Thus, statistics has become an essential part of the pipeline leading to the correct physical interpretation.

The complexity of the instruments, of the astronomical sources, and of the scientific questions lead to many subtle inference problems that require sophisticated statistical tools. For example, data are typically subject to non-uniform stochastic censoring, heteroscedastic errors in measurement, and background contamination. Scientists wish to draw conclusions as to the physical environment and structure of the source, the processes and laws which govern the birth and death of planets, stars, and galaxies, and ultimately the structure and evolution of the universe. Sophisticated physics-based computer-models are used along with complex parametrised and/or flexible multi-scale models to predict the data observed from astronomical sources and populations of sources. Drawing inference under these sophisticated models using complex and/or massive astronomical data streams requires the establishment of whole new statistical frameworks and relies on innovative computational methods. In this way, astrostatistics is pushing the boundaries of both statistics and astrophysics, and is giving rise to entirely new approaches for dealing with complex data sets and uncertainty. Indeed, the well-established physics-based models, existence of carefully quantified measurement errors, and opportunities for cross-validation afforded by massive overlapping data streams make astronomy an ideal test bed for prototyping the modern statistical methods needed in today’s data-rich computationally intensive scientific environment. Thus, astrostatistics should not be viewed as only devoted to the development of new methods for astronomy, but also as an opportunity to establish new general statistical methods, especially in signal processing, data mining, statistical learning, multilevel modelling, and computational statistics.