Created
September 21, 2015 23:44
-
-
Save davidrichards/3ced3fe28e7266eda899 to your computer and use it in GitHub Desktop.
Some Python code I did for a blog article.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import scipy.stats as st | |
import numpy as np | |
import pandas as pd | |
import matplotlib.pyplot as plt | |
def z_score(x, m, s): | |
return (x - m) / s | |
def p(x, m, s): | |
z = z_score(x, m, s) | |
return st.norm.cdf(z) | |
def main(): | |
# z-scores | |
print(z_score(95, 100, 15), z_score(130, 100, 15), z_score(7, 100, 15)) | |
# We should see -0.3333333333333333 2.0 -6.2 or 1/3 deviation below average, 2 above and 6.2 below. | |
# probabilities | |
scores = np.arange(60, 161, 20) | |
z_scores = list(map(lambda x: z_score(x, 100, 15), scores)) | |
less_intelligent = list(map(lambda x: p(x, 100, 15), scores)) | |
df = pd.DataFrame() | |
df['test_score'] = scores | |
df['z_score'] = z_scores | |
df['less_intelligent'] = less_intelligent | |
print(df) | |
# visualization | |
get_ipython().magic('matplotlib inline') | |
mu, sigma = 100, 15. # mean and standard deviation | |
s = sorted(np.random.normal(mu, sigma, 1000)) | |
count, bins, ignored = plt.hist(s, 30, normed=True) | |
plt.style.use('ggplot') | |
plt.plot(bins, 1/(sigma * np.sqrt(2 * np.pi)) * | |
np.exp( - (bins - mu)**2 / (2 * sigma**2) ), | |
linewidth=2) | |
plt.show() | |
# IQ Score of 7 | |
z = z_score(7, 100, 15) | |
prob = p(7, 100, 15) | |
rounded_prob = round(prob, 15) | |
print("The z-score {0} and probability {1} of a test score of 7.".format(z, rounded_prob)) | |
instances_per_billion = round((1/prob) / 1000000000, 2) | |
people_on_the_planet = 7.125 # billion | |
instances_on_the_planet = people_on_the_planet / instances_per_billion | |
instances_on_the_planet | |
# Likely Voters | |
votes = pd.Series([46.3, 45.3, 46.3, 46.3, 49.4, 47.8, 42.7, 43.3, 49.0, 47.7, 48.3, 46.5, 46.5, 49.0, 48.0]) | |
# I thought it was easier to read percentages as 46.3, but I'm converting those numbers here to fit | |
# in the set [0,1] as well-behaved probabilities do. | |
votes = votes.apply(lambda x: x / 100) | |
votes.describe() | |
hillary_z_score = st.norm.ppf(votes.mean()) | |
hillary_z_score | |
iq = 15 * hillary_z_score + 100 | |
iq | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment