Hi !
I create this topic to introduce ourselves. I start :
I am 28 and work as a Financial Risk Manager in France. I have got a degree in finance but did not know much about Bayesian statistics in college.
I am very interested in Bayesian Statistics because it seems more straightforward to me. My goal in the next coming months is to be able to rethink what I do from a Bayesian perspective.
On a daily basis I model Portfolios to compute Value at Risks, volatilities, (Ex-Ante) Tracking-Errorsā¦
I will do my best to bring my knowledge on these subjects !
I am Junpeng Lao, a Post-doc in psychology, currently in University of Fribourg, Switzerland. So if you are interested in open science and how to deal with replication crisis using Bayesian Statistics, please drop me a message.
I learned Python and Bayesian Statistics by learning PyMC3. As part of my learning process, I ported the Lee and Wagenmakersā Bayesian Cognitive Modeling book into PyMC3 (repository here). Currently, I am part of pymc_devs and contribute to the code wherever I can
I also help set up the pymc discourse community, so if you have any suggestion, please let me know!
Thanks @junpenglao for building this community ! I think this is a good initiative to help people getting started with Bayesian analysis in general and PyMC3 in particular.
Iām Maxim, a bachelor student from Moscow State University, I study Economics there. In spare time I learn Approximate Bayesian Inference and itās applications. Currently I have an internship at Yandex as Data Scientist. I also love active sports especially roller-skating.
Iām Thomas, PyMC3 developer and Director of Data Science at Quantopian. I did my PhD on computational psychiatry at Brown where I got exposed to Bayesian statistics and PyMC2. At Quantopian we use PyMC3 to evaluate trading algorithms and track uncertainty over time.
Iām Jon, I work as a data scientist for a subscription/ppv video service in Scandinavia (viaplay.com). Iām from a design/product development background so compared to most of you guys Iām very much a noob. My work mainly consists of regression modeling, and divergent transitions are my mortal enemy =).
Hi, Iām Austin. Iām a PyMC3 developer (mostly documentation and examples) and Principal Data Scientist at Monetate. I dropped out of a PhD program in pure math at the University of Illinois and now I build machine learning products for marketers. I love PyMC3 because it has been, at least for me, the shortest path between math idea and Bayesian inference. I hate hyperparameter optimization, and I am fascinated by the ability of Bayesian nonparametric models to choose what are normally thought of as hyperparameters.
Hi, Iām Jonathan. Iām a quant/trader. I recently left WorldQuant, and will soon be starting a new position at a prop firm in Austin, TX. I really like the intuitiveness of probabilistic programming and the simplicity of PyMC3.
Hey! Iām Colin. Iām a PyMC3 developer (main contributions have been in the Hamiltonian methods, but also improving testing/CI processes) and software engineer in Cambridge, MA. In my past I was a pure mathematician who studied geometric measure theory. I love working on pymc3 because of the pleasant interaction between theory and application ā I am super excited about how the library is catching up with state of the art algorithms, and how friendly the community is!
Hi Iām Peadar Coyle, I work as a Data Scientist for ElevateDirect. Iāve a Masters in Mathematics from the University of Luxembourg (and it was just shortly after that period when I met Thomas Wiecki).
Iām a contributor to PyMC3 (mostly documentation, examples and some model evaluation stuff) - I use it professionally for things like survival analysis of customers.
Hello! Iām Bhargav, I am finishing up my undergrad thesis in INRIA, France on Machine Learning stuff. Iām a Google Summer of Code student for the summer, working on RMHMC. I love the python data science community, and like to travel around to different PyCons and PyDatas in Europe to talk about the awesome python data science stack.
Hello,
Iām Marco and I am a Ph.D. student in computer science at University of Trento. I study how new sources of data (e.g. mobile phone data) can be useful to study the behaviour of cities through data mining
Iām Osvaldo. I am a researcher at The National Scientific and Technical Research Council (Argentina), where I work in the field of structural bioinformatics. I am trying to, increasingly, devote most of my work-time to probabilistic modeling of biomacromolecular structures (using PyMC3, of course!).
I also contribute code to PyMC3 whenever I can, and I have helped to make the code from the puppy book and Statistical rethinking available in Python/PymC3. Learning Bayesian statistics was a game changer for me, suddenly many things began to make sense! I love Bayesian statistics and PyMC3 and I want to share my love with others, now I am preparing material for an introductory Bayesian course, that will be freely available (but probably in Spanish). I am helping to host the first Latin American PyData, and one of the keynote speakers will be Chris Fonnesbeck!
I also love other things like music and musical instruments (specially bass and guitar), drinking beers, movies, bikingā¦ sometimes I think I will need more or less 5 lifetimes to do all that I want and then I realize sampling is really fast with PyMC3 and I feel hope again!
Hi, Iām Ben
Iām a lecturer in a small psychology department in Scotland, UK (lab website). I guess Iām a domain specialist in that Iām a cognitive scientist, but Iāve also been Bayesian for a while and done a fair bit of modelling of some sort or another.
Iāve done a bit of work on Bayesian models of visual attention. More recently, Iāve switched over to looking at āhigher-levelā decision making. Specifically Iām interested in risky and delayed choice tasks, basically decision making.
Iāve got a long history with Matlab, and I recently published a Toolbox which does hierarchical inference for people with delay discounting data. Iāve got many plans to extend this any make it super awesome - but itās become clear that thatās not possible in Matlab/JAGS. So Iām trying to port this over to Python/PyMC3. At the moment, itās in stealth mode, but I might open up the GitHub repo and welcome some co-development in exchange for co-authorship
While Iām a Bayesian, and I understand the simpler sampling algorithms, Iām not a computer scientist nor a statistician. So itās unlikely that I can contribute much to core PyMC3, but Iām open to contributing some examples once Iām up to speed.
Iām a researcher at Yale University. I study hearing, quantitative genetics, and computational psychiatry, whatever that is. You can read more about me here if you like.
Iām more of an enthusiastic consumer than contributor to PyMC3. My first paper using PyMC3 just got accepted and should be online soon!
This sounds fascinating! I might be interested in helping port this sort of thing to PyMC3ā¦do you have any relevant references I could read to ground myself in the basic psychology, but moreso the relevant mathematical models (I am a mathematician after all).
:
