probability distribution function
ASU W.P. Carey School of Business
Arizona State University
Course: Data Distribution
Instructor: Dr. Asish Satpathy
Learning Goals
Distribution of continuous variables
Distribution of discrete variables
Random Variables
Probability Distribution of Random Variables
Description for Continuous Variables
Continuous Data Variables: Used to describe data that can take any value within a given range.
Centrality Metrics:
Mean
Median
Mode
Spread Metrics:
Quartiles
Range
Standard Deviation
Variance
Interquartile Range
Mean Absolute Deviation
Shape Indicators:
Skewness
Kurtosis
Distribution of Continuous Variable
Mean: -0.246661
Standard Deviation: 1.0010272
Number of Observations (N): 50
Variance: 1.0020554
Skewness: -0.154412
Kurtosis: -0.345945
Range: 4.345398
Description for Discrete Variables
Discrete Data Variables: Represent data that can only take certain distinct values.
Centrality Metrics:
Mode (derived from frequency graphs, but not particularly meaningful)
Shape: Defined by the distribution of frequencies.
Distribution of Discrete Variables
Frequencies:
Impressionism: Count - 13, Probability - 0.23636
Landscape: Count - 13, Probability - 0.23636
Modern: Count - 15, Probability - 0.27273
Performance: Count - 8, Probability - 0.14545
Renaissance: Count - 6, Probability - 0.10909
Total Count: 55, Total Probability: 1.00000
Random Variables
Continuous Random Variable Examples:
Amount of rain in Tempe in the next month.
Length of time waiting for a table at a restaurant.
Discrete Random Variable Examples:
Number of bounced checks on a given day at the bank.
Number of drinks ordered by first customers at a Starbucks.
Probability Distribution for Continuous Random Variables
Describes the probability associated with continuous variables.
Graph insights:
Represents chance (not finite data).
Probability for a customer to wait between A and B minutes is the area under the curve.
Waiting time probability from 0 to 60 minutes equals 1 (total area under curve).
Probability Distribution for Discrete Random Variables
Visualized using a probability histogram.
Example for number of heads after three flips of a fair coin:
Outcomes and probabilities are as follows:
0 Heads: Probability = 1/8
1 Head: Probability = 3/8
2 Heads: Probability = 3/8
3 Heads: Probability = 1/8
Normal and t-Distribution Functions
Graphical Representation of normal distribution function.
Depicts the shape and probabilities related to a standard normal distribution.
t-distribution is used when n is close to or smaller than 30.
Poisson Distribution (Non-normal)
Describes the probability of a given number of events occurring in a fixed interval of time or space.
Related probabilities depicted on a graphical function.
Test Your Knowledge
Question 1: Which of the following measures is not used in describing a continuous data variable?
A: Mean
B: Variance
C: Frequency
D: Kurtosis
Question 2: Which of the following is an example of a discrete random variable?
A: Number of children in a family
B: The amount of sugar in an orange
C: Time required to run a five-mile marathon
D: Profit margin in the next quarter