Bayesian statistics's topics - tribe.net

Multiple testing

Thu, 12 Jan 2006 21:10:14 GMT

does anyone know any good ways to correct for multiple testing? I'll be more specific: I have a million hypotheses many of which may simultaneously be correct. I want to find the most significant predictions. however, with a p-value threshold of .05, there should be 50 wrong predictions in those with p-value < .05. how might I set a more appropriate p-value cutoff?

posted in Bayesian statistics - 2 replies

priors

Sat, 07 Jan 2006 00:54:37 GMT

if a friend has gotten absolutely rat-assed on wine every time I've been over to his house for dinner, what is the posterior probability that he will again do so tonight?

(keep in mind that it's the beginning of a new year and sometimes people make resolutions about such things...)

posted in Bayesian statistics - 2 replies

BUGS beginner

Wed, 28 Sep 2005 01:16:45 GMT

Anyone out there using BUGS?

I'm just starting out...and could do with links to tutorial / introductions with a bit more explanation that the included examples.

Can anyone help, please?

posted in Bayesian statistics - 5 replies

note from tribe.net: please assign a moderator

touguy — Fri, 06 May 2005 20:11:19 GMT

Hey there, "Bayesian statistics" members--

This Tribe doesn't have a moderator, but as a matter of policy, Tribe.net likes all groups to have a leader.
Could you collectively choose someone to moderate your Tribe?

When you've reached consensus, have the new moderator send a note to help@tribe.net, letting us know that you've selected a new leader.

Thanks,
-- TOU (Terms of Use) Guy

posted in Bayesian statistics - 2 replies

Looking for Natural Language Processing, Machine Learning, etc. folks in Pasadena

gb — Fri, 19 Aug 2005 19:00:23 GMT

First an apology and a promis that I'll remove this posting if its wrong to post this type of message here.

We are a Pasadena-based start-up (and very well funded I might ad) and are in the Search Engine Marketing (SEM) arena. In short we have combined a semantic search engine, a machine learning system and a web-based interface to drive SEM as no other company does it. To be honest, it is far easier to describe the solution on a phone call.

We are building a team of truly outstanding team of professionals. Our team so far includes a winner of Google's CodeJam, a multiple participant in international algorithmic competitions, and a few alumni of Caltech, Stanford, & CAL. In short, this is a remarkable team of SW Engineers. Our criteria for selection is pretty high, and they include:

Brilliant & can get things done - Being smart is fine, but being smart and getting things done right the first time is better.

Problem Solvers - We have some complex problems to solve, so everyone needs to provide multiple ways to possibly solve them.

Good Person - We are not just about working. We want to be with people we enjoy sitting down to a meal with, want to hear about the family, and be a friend and a colleague.

A true ‘coder’ – We like the type of person that if stranded in a cave would be writing code on the walls with burnt sticks.

So, if you feel you are up to the challenge drop me a note or a call and lets talk.

Best regards,

gb

Greg Buechler
greg@adapt.com
925-968-1737

Monday, August 1, 2005
Adapt Technologies Raises $9.8M

Businessweek is reporting this week that Pasadena-based MarketFinder has raised $9.8M in venture funding. The funding came from Mohr Davidow Ventures, Emergence Capital Partners, and Concept2company. The firm develops software that automates the placement and management of keyword-based, CPC ads in the paid search space. The firm is headed by Alex Backer, a visiting scientist at the California Institute of Technology.
posted on Monday, August 1, 2005

posted in Bayesian statistics - 0 replies

in a nutshell

Ann — Fri, 28 Jan 2005 04:28:12 GMT

If you had to define Bayesian statistics in 100 words or less, what would you say?

posted in Bayesian statistics - 3 replies