Normal distributions
F O U N D AT I O N S O F P R O B A B I L I T Y I N P Y T H O N
Alexander A. Ramírez M.
CEO @ Synergy Vision
Modeling for measures
FOUNDATIONS OF PROBABILITY IN PYTHON
Adults' heights example
FOUNDATIONS OF PROBABILITY IN PYTHON
Probability density
FOUNDATIONS OF PROBABILITY IN PYTHON
Probability density examples
FOUNDATIONS OF PROBABILITY IN PYTHON
Probability density and probability
FOUNDATIONS OF PROBABILITY IN PYTHON
Symmetry
FOUNDATIONS OF PROBABILITY IN PYTHON
Mean
FOUNDATIONS OF PROBABILITY IN PYTHON
Mean (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
Mean (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
Standard deviation
FOUNDATIONS OF PROBABILITY IN PYTHON
Standard deviation (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
Standard deviation (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
One standard deviation
FOUNDATIONS OF PROBABILITY IN PYTHON
Two standard deviations
FOUNDATIONS OF PROBABILITY IN PYTHON
Three standard deviations
FOUNDATIONS OF PROBABILITY IN PYTHON
Normal sampling
# Import norm, matplotlib.pyplot, and seaborn
from scipy.stats import norm
import matplotlib.pyplot as plt
import seaborn as sns
# Create the sample using norm.rvs()
sample = norm.rvs(loc=0, scale=1, size=10000, random_state=13)
# Plot the sample
sns.distplot(sample)
plt.show()
FOUNDATIONS OF PROBABILITY IN PYTHON
Normal sampling (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
Let's do some
exercises with
normal distributions
F O U N D AT I O N S O F P R O B A B I L I T Y I N P Y T H O N
Normal probabilities
F O U N D AT I O N S O F P R O B A B I L I T Y I N P Y T H O N
Alexander A. Ramírez M.
CEO @ Synergy Vision
Probability density
In Python this can be done in a couple of
lines:
# Import norm
from scipy.stats import norm
# Calculate the probability density
# with pdf
norm.pdf(-1, loc=0, scale=1)
0.24197072451914337
loc parameter speci es the mean and
scale parameter speci es the standard
deviation.
FOUNDATIONS OF PROBABILITY IN PYTHON
pdf() vs. cdf()
FOUNDATIONS OF PROBABILITY IN PYTHON
pdf() vs. cdf() (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
pdf() vs. cdf() (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
Cumulative distribution function examples
# Calculate cdf of -1 # Calculate cdf of 0.5
norm.cdf(-1) norm.cdf(0.5)
0.15865525393145707 0.6914624612740131
FOUNDATIONS OF PROBABILITY IN PYTHON
The percent point function (ppf)
# Calculate ppf of 0.2 # Calculate ppf of 55%
norm.ppf(0.2) norm.ppf(0.55)
-0.8416212335729142 0.12566134685507416
FOUNDATIONS OF PROBABILITY IN PYTHON
ppf() is the inverse of cdf()
# Calculate cdf of value 0 # Calculate ppf of probability 50%
norm.cdf(0) norm.ppf(0.5)
0.5 0
FOUNDATIONS OF PROBABILITY IN PYTHON
Probability between two values
# Create our variables
a = -1
b = 1
# Calculate the probability between
# two values, subtracting
norm.cdf(b) - norm.cdf(a)
0.6826894921370859
FOUNDATIONS OF PROBABILITY IN PYTHON
Tail probability
# Create our variable
a = 1
# Calculate the complement
# of cdf() using sf()
norm.sf(a)
0.15865525393145707
FOUNDATIONS OF PROBABILITY IN PYTHON
Tails
# Create our variables
a = -2
b = 2
# Calculate tail probability
# by adding each tail
norm.cdf(a) + norm.sf(b)
0.04550026389635839
FOUNDATIONS OF PROBABILITY IN PYTHON
Tails (Cont.)
# Create our variables
a = -2
b = 2
# Calculate tail probability
# by adding each tail
norm.cdf(a) + norm.sf(b)
0.04550026389635839
FOUNDATIONS OF PROBABILITY IN PYTHON
Intervals
# Create our variable
alpha = 0.95
# Calculate the interval
norm.interval(alpha)
(-1.959963984540054, 1.959963984540054)
FOUNDATIONS OF PROBABILITY IN PYTHON
On to some
practice!
F O U N D AT I O N S O F P R O B A B I L I T Y I N P Y T H O N
Poisson distributions
F O U N D AT I O N S O F P R O B A B I L I T Y I N P Y T H O N
Alexander A. Ramírez M.
CEO @ Synergy Vision
Poisson modeling
FOUNDATIONS OF PROBABILITY IN PYTHON
Poisson distribution properties
FOUNDATIONS OF PROBABILITY IN PYTHON
Probability mass function (pmf)
Imagine you have 2.2 calls per minute.
FOUNDATIONS OF PROBABILITY IN PYTHON
Probability mass function (pmf) (Cont.)
In Python we do the following:
# Import poisson
from scipy.stats import poisson
# Calculate the probability mass
# with pmf
poisson.pmf(k=3, mu=2.2)
0.19663867170702193
mu parameter speci es the mean of
successful events.
FOUNDATIONS OF PROBABILITY IN PYTHON
pmf examples
# Calculate pmf of 0 # Calculate pmf of 6
poisson.pmf(k=0, mu=2.2) poisson.pmf(k=6, mu=2.2)
0.11080315836233387 0.01744840480280308
FOUNDATIONS OF PROBABILITY IN PYTHON
Different means
FOUNDATIONS OF PROBABILITY IN PYTHON
Cumulative distribution function (cdf)
# Calculate cdf of 2 # Calculate cdf of 5
poisson.cdf(k=2, mu=2.2) poisson.cdf(k=5, mu=2.2)
0.6227137499963162 0.9750902496952996
FOUNDATIONS OF PROBABILITY IN PYTHON
Survival function and percent point function (ppf)
# Calculate sf of 2 # Calculate ppf of 0.5
poisson.sf(k=2, mu=2.2) poisson.ppf(q=0.5, mu=2.2)
0.3772862500036838 2.0
FOUNDATIONS OF PROBABILITY IN PYTHON
Sample generation (rvs)
# Import poisson, matplotlib.pyplot, and seaborn
from scipy.stats import poisson
import matplotlib.pyplot as plt
import seaborn as sns
# Create the sample using poisson.rvs()
sample = poisson.rvs(mu=2.2, size=10000, random_state=13)
# Plot the sample
sns.distplot(sample, kde=False)
plt.show()
FOUNDATIONS OF PROBABILITY IN PYTHON
Sample generation (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
Let's practice with
Poisson
F O U N D AT I O N S O F P R O B A B I L I T Y I N P Y T H O N
Geometric
distributions
F O U N D AT I O N S O F P R O B A B I L I T Y I N P Y T H O N
Alexander A. Ramírez M.
CEO @ Synergy Vision
Geometric modeling
FOUNDATIONS OF PROBABILITY IN PYTHON
Geometric parameter
Model for a basketball player with probability We can model a grizzly bear that has a 0.033
0.3 of scoring. probability of catching a salmon.
FOUNDATIONS OF PROBABILITY IN PYTHON
Probability mass function (pmf)
In Python we code this as follows:
# Import geom
from scipy.stats import geom
# Calculate the probability mass
# with pmf
geom.pmf(k=30, p=0.0333)
0.02455102908739612
p parameter speci es probability of success.
FOUNDATIONS OF PROBABILITY IN PYTHON
Cumulative distribution function (cdf)
# Calculate cdf of 4
geom.cdf(k=4, p=0.3)
0.7598999999999999
FOUNDATIONS OF PROBABILITY IN PYTHON
Survival function (sf)
# Calculate sf of 2
geom.sf(k=2, p=0.3)
0.49000000000000005
FOUNDATIONS OF PROBABILITY IN PYTHON
Percent point function (ppf)
# Calculate ppf of 0.6
geom.ppf(q=0.6, p=0.3)
3.0
FOUNDATIONS OF PROBABILITY IN PYTHON
Sample generation (rvs)
# Import poisson, matplotlib.pyplot, and seaborn
from scipy.stats import geom
import matplotlib.pyplot as plt
import seaborn as sns
# Create the sample using geom.rvs()
sample = geom.rvs(p=0.3, size=10000, random_state=13)
# Plot the sample
sns.distplot(sample, bins = np.linspace(0,20,21), kde=False)
plt.show()
FOUNDATIONS OF PROBABILITY IN PYTHON
Sample generation (rvs) (Cont.)
FOUNDATIONS OF PROBABILITY IN PYTHON
Let's go try until we
succeed!
F O U N D AT I O N S O F P R O B A B I L I T Y I N P Y T H O N
