deanmalmgren

"""Pipe the output of this script to another program to look at the CDF of the values. For example,

python random_numbers.py 1000 | sort -n | awk '{print $1, 1-(NR-1)/1000}' | xmgrace -
"""

import sys
import random

n = int(sys.argv[1])
for i in xrange(n):

deanmalmgren

sample_a.dat
sample_b.dat

deanmalmgren

sample_a.dat
sample_b.dat

deanmalmgren

When I saw this on twitter, it really made me wonder whether the reported statistics had any significance. Based on the numbers reported, I wrote a quick little program to estimate the 95% confidence interval of the percent of CEOs hired internally. Alongside each regions name is the lower and upper cutoff on the 95% confidence interval.

[unix]$ python simulate.py
          USA/Canada	0.679	0.859
      Western Europe	0.625	0.854
               Japan	0.897	1.000
        Other mature	0.565	0.826
               China	0.710	0.968
Brazil, Russia, India	0.517	0.828

deanmalmgren

This gist has some quick analysis related to the outcomes from an elongated dice experiment. analyze_roll_data.py is a script to conduct the analysis. p16_aspect_ratio.agr contains the results of this analysis in an xmgrace file.

We tried a few quick things to work out a theoretical relationship between the

deanmalmgren

This is a quick visualization of the Scialog pre-conference network. Orange nodes are theorists, blue nodes are experimentalists, and yellow circles are both experimentalists and theorists. For simplicity, this only displays the bidirectional links between pairs of scientists where the thickness of the line joining two scientists corresponds with how strongly they know each other.

To run this in development, just use python's simple server:

python -m SimpleHTTPServer

deanmalmgren

1.0 0.022566380334 0.04 0.0245276407981 0.029769291226 0.0536675041929 0.0268831375565 0.0421124148346 0.0320871215252 0.0444774273755 0.0444774273755 0.0186396379746 0.0363831094317 0.029769291226 0.0105892274179 0.0348216114596 0.0134713615846 0.0194822293602 0.0320871215252 0.029769291226 0.0363831094317 0.0308824682094 0.0444774273755 0.0682274642961 0.021983276081 0.0277777777778 0.0421124148346 0.0444774273755 0.0238353731866 0.0171625207348 0.0381007372999 0.0320871215252 0.0209064363614 0.0363831094317 0.0320871215252 0.0381007372999 0.0231827458604 0.0471451211741 0.04 0.021983276081 0.0238353731866 0.029769291226 0.0209064363614 0.0260468863234 0.00918543875722 0.00721809760087 0.0348216114596 0.0252633672918 0.0138953290753 0.029769291226
0.022566380334 1.0 0.0252633672918 0.029769291226 0.0141178336851 0.0182460947032 0.0287373176349 0.0471451211741 0.0348216114596 0.0214307586872 0.0333953686704 0.0186396379746 0.0363831094317 0.0231827458604 0.0130732739486 0.0231827458604 0.0333953686704 0.0333

deanmalmgren

# from the bash prompt...

# starting from the directory that contains repositories X and Y, use 'hg convert --filemap' to find 
# extract the content from path/to/dir in repository X and put it into a temporary repository TMP
echo 'include "path/to/dir"' > filemap.txt
hg convert --filemap.txt ./X ./TMP

# enter repository Y and do a "force pull" to pull in the changes from the temporary repository TMP
cd ./Y
hg pull -f ../TMP

deanmalmgren

This is a simple gist to illustrate the peculiar behavior with django 1.5 and DEBUG=False when running locally (and in production?...haven't gotten that far yet).

deanmalmgren

python tools/install_paths \
  	libdir=/usr/local/Cellar/meld/1.7.2/lib/meld \
		localedir=/usr/local/Cellar/meld/1.7.2/share/locale \
		helpdir=/usr/local/Cellar/meld/1.7.2/share/gnome/help/meld \
		sharedir=/usr/local/Cellar/meld/1.7.2/share/meld \
		< bin/meld > bin/meld.install
python tools/install_paths \
		libdir=/usr/local/Cellar/meld/1.7.2/lib/meld \
		localedir=/usr/local/Cellar/meld/1.7.2/share/locale \
		helpdir=/usr/local/Cellar/meld/1.7.2/share/gnome/help/meld \