Probability Distributions¶
Some Common Distributions¶
Normal (and Log-Normal) Distribution¶
The normal distributions are used to explain central tendencies
Defined using two parameters
Mean (\(\mu\)) and Standard Deviation (\(\sigma\))
Symmetric distribution
Represents additive processes
The lognormal distribution stipulates that the log of random variable x (i.e., log(x) is normally distributed
Does not work for negative data (in orginal space)
Represents multiplicative process
Lognormal distribution is often appropriate when the data are skewed. (See the Gamma distribution later on) Normal and Lognormal Distributions are best suited to represent central tendencies
Python Examples¶
Poisson Distribution¶
A discrete distribution used to represent the number of independent events within a fixed time
Number of independent rainfall events within a year
Related to Exponential Distribution
Continuous distribution for inter-arrival times
Poisson distribution assumes stationarity
Rate at which events occur is constant
An example of application is Synthetic Rainfall Generation Model for Evaluating Potential Erosion at Highway Construction Sites
Python Examples¶
Exponential Distribution¶
Often used to model the time between two independent events
Time between two rainstorms
Exponential distribution is represented by the parameter (\(\lambda\))
Reciprocal of Average inter-event time
Related to Poisson Distribution
A discrete distribution for number of events in a fixed time
Python Examples¶
Binomial Distribution¶
Used when there are two outcomes
Success and Failure
The probability of success for each event is denoted by “p”
Often assumed stationary
The PMF calculates the probability of success of x out of N total events
The binomial distribution is often used for risk and reliability calculations
Python Examples¶
References¶
Chan, Jamie. Machine Learning With Python For Beginners: A Step-By-Step Guide with Hands-On Projects (Learn Coding Fast with Hands-On Project Book 7) (p. 2). Kindle Edition.
Machine Learning for CE Probability Distributions (Fall 2020 Lesson)
Synthetic Rainfall Generation Model for Evaluating Potential Erosion at Highway Construction Sites