Chris Geoga

I am a statistician and applied mathematician with broad interests in statistical methodology and computing in the physical sciences. My primary interest is modeling complicated dependence structure in real-life temporal/spatial/spatio-temporal processes, and I'm particularly interested in theoretical questions that are motivated by computationally scalable methods and approximations.

Education:

Ph.D. Statistics, Rutgers University (2018 – 2023)

Dissertation: Models and Computation for Nonstationary Processes at Scale
Adv. Michael L. Stein

B.S. Mathematics / B.A. Statistics, University of Chicago (2012 – 2016)

Employment:

University of Wisconsin-Madison, Dept. of Statistics

Assistant Professor (2023 – )

Argonne National Laboratory, Division of Mathematics and Computer Science

Visiting Student (2022 – 2023)
Argonne Associate (2018 – 2022)
Assistant Computational Mathematician (RD1) (2016 – 2018)

Publications:

Clicking on the title of the paper will take you to ArXiv or some other direct PDF source, and the DOI link will take you to the official journal version. Direct PDF links marked (ArXiv vXXX) mean that the ArXiv version differs from the journal version somehow. I try to do a good job about uploading substantial revisions to arXiv, so the differences are in general limited to small revisions that don't really change the substance of the paper very much. But if you want a journal version and can't get access, please shoot me a note and I'll send one over.

You can also see my publications on My google scholar profile. Authors in maroon are/were graduate students working with me at the time of initial submission.

H. Xiong, P. Muradyn, and C.J. Geoga. High-Resolution Retrieval of Atmsopheric Boundary Layers with Nonstationary Gaussian Processes. Submitted.
C.J. Geoga, P.G. Beckman. Fast nonparametric spectral density estimation from irregularly sampled data. Submitted.
P.G. Beckman, C.J. Geoga. Fast Adaptive Fourier Integration for Spectral Densities of Gaussian Processes. Statistics and Computing. DOI link.
C.J. Geoga. Fast Machine-Precision Spectral Likelihoods for Stationary Time Series, SISC. DOI link.
C.J. Geoga, M.L. Stein. A Scalable Method to Exploit Screening in Gaussian Process Models with Noise (Arxiv v2). Journal of Computational and Graphical Statistics. DOI link.
P.G. Beckman, C.J. Geoga, M.L. Stein, and M. Anitescu. Scalable Computations for Nonstationary Gaussian Processes (Arxiv v1). Statistics and Computing. DOI link.
C.J. Geoga, O. Marin, M. Schanen, and M.L. Stein. Fitting Matern smoothness parameters using automatic differentation . Statistics and Computing. DOI link.
C.J. Geoga, M. Anitescu, and M.L. Stein. Flexible nonstationary spatio-temporal modeling of high-frequency monitoring data (ArXiv v1). Environmetrics. DOI link.
C.J. Geoga, M. Anitescu, and M.L. Stein. Scalable Gaussian process computations using hierarchical matrices. Journal of Computational and Graphical Statistics. DOI link.
C.J. Geoga, C.L. Haley, A. Siegel, and M. Anitescu. Frequency-wavenumber spectral analysis of spatiotemporal flows. Journal of Fluid Mechanics. DOI link.

Software:

This is a selection of my most polished and (in my opinion) useful software projects. See my Github and Sourcehut profiles for other projects and things in development. All code is written in the Julia language. If you are an R user but want to use any of this software, feel free to reach out---I have provided or helped other people develop bindings for use in R before. It isn't too bad to do.

Vecchia.jl, A very optimized library for fitting Gaussian process models with Vecchia's approximation. This library offers a lot of features that no other library does: chunking (where the prediction sets are not singletons), complete autodiff compatibility (optimize with real Hessians!), arbitrary mean and covariance functions, and so on. The core package has very few dependencies to keep things light, but I have also implemented several extensions so make it trivial to hook into advanced optimizers and bring your own exotic specifications of the approximation and so on. Give it a try!
BesselK.jl, An implementation of the modified second-kind Bessel function K_v that is `ForwardDiff.jl`-compatible with respect to the order. See the paper "Fitting Matern..." for more details on this. This is the only software library in the world (to my knowledge) that gives you those derivatives to machine precision (and fast!). I am in the process of porting this to C and making it available as R and python packages as well.
IrregularSpectra.jl, The software companion to "Nonparametric spectral...". It provides functionality where you give it fully arbitrary point locations and measurements and you get back a good estimate for the spectral density. Like the bessel library, this is the only software in the world (to my knowledge) that gives you a decent estimator for irregular point measurement patterns that aren't just a gappy lattice---and to provably do it in quasilinear time! And if you are on a gappy lattice, I would bet a beer that this library is faster than any competitors. But with that said, the software design problem here is quite hard, and so if you try it and aren't happy with your results, I would appreciate you opening an issue before dismissing it.

Contact:

Besides for a Mastodon account that I don't check, I do not use any social media or internet networking websites. If you see any account with my name on it, please assume that it is not really me. Please contact me by email at the following address:

geoga $at wisc $dot edu

I also use XMPP, so if you'd prefer to communicate by instant messaging please first reach out by email and I will share that contact information with you.

                    
                       `
                             `      
                             `      
            -+oss++o+osso+/::o-`````
          :oosssyyo/+sshhhhyyy-   
        :ossyhhysso/:--:///:-.``      
       +syyysys+-`                  
      +yhyss+-                      
     -syss/`                        
     :syy`                          
    //sy-                           
   `+ooo                            
    :/-