2029 search results for "Twitter"

Adapting graphs for presentations

October 28, 2010
By
Adapting graphs for presentations

I’ve just finished reading slide:ology by Nancy Duarte. It contains lots of advice about how to convey meaning through aesthetics. The book has a general/business presentation focus, but it got me wondering about how to apply the ideas in a scientific context.  Since graphs from a big part of most scientific talks, and since that’s

Read more »

Random generators for parallel processing

October 28, 2010
By
Random generators for parallel processing

Given the growing interest in parallel processing through GPUs or multiple processors, there is a clear need for a proper use of (uniform) random number generators in this environment. We were discussing the issue yesterday with Jean-Michel Marin and briefly looked at a few solutions: given p parallel streams/threads/processors, starting each generator with a random

Read more »

Where People Share Links About NYC

October 27, 2010
By
Where People Share Links About NYC

Last week I participated in bit.ly’s fourth hackabit hack-a-thon, which is a wonderful opportunity for NYC area hackers to get together, eat pizza, drink energy drinks, and stay up late hacking with some of the best data geeks around. I was lucky enough to saddle up next to Hilary Mason, bit.ly’s lead scientist, recently named

Read more »

Parametric Bootstrap Power Analysis of GISS Temp Data

October 24, 2010
By
Parametric Bootstrap Power Analysis of GISS Temp Data

Previosly, I calculated a bunch of ad-hoc power curves from GISTEMP data. Power is essentially a reframing of the p-value, to see the significance of the trend lines in the global temps. However, power calculations are inherently very noisy, hence, my ad-hoc way of aggregating the data. Another method is to bootstrap through the responses

Read more »

Accessing R from Python using RPy2

October 24, 2010
By

This past Tuesday I had the opportunity to present a short talk (a bit long) related to text mining at the Los Angeles R Users’ Group. Since I do most of my text mining in Python, I took this opportunity to discuss RPy2, an interface to R from Python. My slides are below: Accessing R from Python...

Read more »

Le Monde puzzle [42]

October 24, 2010
By
Le Monde puzzle [42]

An interesting suduko-like puzzle for this week puzzle in Le Monde thi A 10×10 grid is filled by a random permutation of {0,…,99}. The 4 largest figures in each row are coloured in yellow and the 4 largest values in each column are coloured in red. What is the range of the number of yellow-and-red

Read more »

Incremental improvements to Nightlights mapping thanks to R-Bloggers

October 22, 2010
By
Incremental improvements to Nightlights mapping thanks to R-Bloggers

The R community is very generous and collaborative. This post walks through the suggestions left by commenters to my previous post on Steve Mosher's Nightlights work, and show the resulting much-improved output.

Read more »

abc

October 21, 2010
By
abc

Michael Blum and Olivier François, along with Katalin Csillery, just released an R package entitled abc. (I am surprised the name was not already registered!) Its aim is obviously to implement ABC approximations for Bayesian inference: Description The ’abc’ package provides various functions for parameter estimation and model selection in an ABC framework. Three main

Read more »

vecLib: Why Mac users are better off with Open Source R

October 21, 2010
By
vecLib: Why Mac users are better off with Open Source R

Thanks to its use of Apple's vecLib for BLAS, LAPACK and related functions, the open source distribution of R performs many matrix calculations 2-10X faster than Revolution Analytics Community edition on Mac OS X systems.

Read more »

What’s that 5km from the station “location”

October 21, 2010
By
What’s that 5km from the station “location”

In our last installment we looked at stations which were pitch black. The case I examined, Middlesboro Kentucky illustrated 1. The station location data used by Hansen2010 has inaccuracies. 2. While the purported station location was pitch dark, nearby within a couple 1/100ths of a degree there were urban lights. What this example illustrated was

Read more »