addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramFill 1linklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

Data Mining Using Bayesian Data Analysis and Python

Thomas Wiecki will present a talk title 'Bayesian Data Analysis with PyMC3'

Details:

Probabilistic Programming allows flexible specification of statistical models to gain insight from data. Estimation of best fitting parameter values, as well as uncertainty in these estimations, can be automated by sampling algorithms like Markov chain Monte Carlo (MCMC). The high interpretability and flexibility of this approach has lead to a huge paradigm shift in scientific fields ranging from Cognitive Science to Data Science and Quantitative Finance.

PyMC3 is a new Python module that features next generation sampling algorithms and an intuitive model specification syntax. The whole code base is written in pure Python and Just-in-time compiled via Theano for speed.

In this talk I will provide an intuitive introduction to Bayesian statistics and how probabilistic models can be specified and estimated using PyMC3.

BIO:
Thomas Wiecki is currently enrolled in the Ph.D. program at Brown University where he investigates the neuronal underpinnings of mental illness using quantitative methods like Bayesian Modeling. He also works as a quantitative researcher for Quantopian Inc where he helps building the worlds' first browser based financial backtesting platform.

Join or login to comment.

  • Thomas

    Thanks everyone for coming, I hope it was informative!

    Here are the slides of the talk:
    https://rawgithub.com/twiecki/pymc3_talk/master/bayesian_pymc3.slides.html#/

    There was actually a minor error that was pointed out during the talk (fixed in the uploaded slides) where the plot actually showed a Beta(2, 2) distribution instead of Beta(1, 1).

    October 16, 2013

    • Thomas

      Hi Rodrigo, Not that I know of, I just came up with that ;). The idea of using a RandomWalk is from the stochastic volatility model.

      1 · October 16, 2013

    • Bobby M.

      hi Thomas, i really enjoyed your talk and i'd be interested in modeling implied volatility surfaces from options prices using these techniques, would you have any recommendations for resources?

      October 22, 2013

  • Daniel G.

    Thanks for presenting, Thomas!

    October 18, 2013

  • Lydia H.

    Was this presentation recorded?

    October 16, 2013

    • Sheamus M.

      Hi Lydia. Generally our talks are not recorded since there is a cost in hiring someone to do a quality recording. We are actively looking for a sponsor to fund this so perhaps in future we will.

      October 16, 2013

  • Leo U.

    Are the slides available anywhere?

    1 · October 16, 2013

    • Sheamus M.

      Leo. I about to send a link out on an email to all members

      October 16, 2013

  • Ramon V.

    Excellent presentation, expanded my knowledge of Bayes use and python.

    1 · October 16, 2013

  • Sheamus M.

    Thanks everyone for attending!

    October 15, 2013

  • Stuart L.

    Will this talk be recorded?

    October 14, 2013

  • Gary

    These seem to be the main line tools for this type of analysis

    October 7, 2013

Our Sponsors

  • Strata + Hadoop World

    20% special discount code UGBOSDM

  • WeWork

    Contact WeWork, mention your BDM Meetup Membership for special discounts

  • Open Data Science

    Signup for ODSC Newsletter. Receive attractive ODSC East Conf. offers!

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy