-
👋 Hi, I’m @NateJudd
-
👀 I’m interested in Mathematical Statistics. In particular, using probability theory and information theory to say things about statistical models and statistical estimation.
-
⭐ My Doctoral Thesis constructed two classes of non-diffusive Wright-Fisher processes and (1) derived exact, computable posterior, posterior predictive and smoothing distribution formulae, and, (2), derived and implemented exact sampling schemes for both classes.
-
📔 You can find part of this work on ArXiv - https://arxiv.org/abs/2604.11363.
-
💻 You can find the code to build a python module that exactly simulates from the subordinated Wright-Fisher process in my "exact-simulation..." repo.
-
🌱 I have recently worked on lower bounding methods, with an eye for heavy-tailed distributions.
-
💻 You can find a flavour of this work in my "tail-bounds" repo.
-
✨ I am working on applying lower bounding method to identify the optimal sample complexity of important statistical estimation problems.
-
💞️ Open to discuss and collaborate
-
📫 How to reach me nathan.judd@oist.jp
PhD in Statistics; PostDoc;
Interested in Statistics, Probability, Stochastic Processes, Machine Learning and Information Theory.
Pinned Loading
-
crime_linkage_model_comparison
crime_linkage_model_comparison PublicA comparison of crime linkage models on simulated data
Python
-
Exact-simulation-of-time-changed-Wright-Fisher-processes
Exact-simulation-of-time-changed-Wright-Fisher-processes PublicExact simulation of subordinated Wright-Fisher processes
C++
-
Global-Slavery-Index-Data-Interrogation
Global-Slavery-Index-Data-Interrogation PublicData analysis of the global slavery index data set
R
-
-
Introduction-to-filtering-subordinated-Wright-Fisher-priors
Introduction-to-filtering-subordinated-Wright-Fisher-priors PublicThe update step for filtering a subordinated Wright-Fisher process
R
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.
