Scribd
Upload
25 views4 pages

4.1 Normal Distribution: Properties

Uploaded by

Jun Ahn
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
25 views4 pages

4.1 Normal Distribution: Properties

Uploaded by

Jun Ahn
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

4.

1 Normal Distribution

Properties
l Very useful for a variety of problems
l Symmetric, unimodal, bell-shaped curve
l Parameters: mean, standard deviation

Z-Score

 

Putting data onto a “standardised” scale

Important Examples
Comparing the approximate percentile of data, each in different groups
1.
2. Compare the Z-scores

How to solve a normal probability


1. Draw a picture of the normal distribution.
2. Calculate the Z-score.
3. Find the probability using the probability table.

68-95-99.7 Rule
4.2. Geometric Distribution

Bernoulli Distribution
Properties
l Random process with only two outcomes: a success or failure
l Probability of a success ∈    

Mean and SD of a Bernoulli distribution


 and   
    
when success takes the value  and failure takes the value 

Geometric Distribution
Describes how many trials it takes to observe a success
-axis: number of trials until a success; -axis: probability

Mean and SD of Distribution


When the probability of succuess is ,
Probability of finding the first success in the  th trial:      
Mean:     

SD:   



4.3 Binomial Distribution

Properties
l Used to describe the number of successes in a fixed number of trials
l A “final probability” is identified:  of scenarios   single scenario
l  single scenario        (for  successes and  trials)

    
 

Mean and SD
   and   
    

Conditions
l Independence
l Fixed number of trials
l Outcome is either success or failure
l  is the same for each trial

Computing Binomial Probabilities


1. Check if the model is appropriate.
2. Identify     and .
3. Use the formula to do the calculations.

Normal Approximation
If  and      are both big enough, at least , then we can approximate the binomial
distribution as a normal distribution.

Steps
1. Check if  and      are greater than or equal to .
2. Compute  and .
3. Use the normal model    to estimate the probability of observing a certain range of
successes.

Bad approximation
Even if the conditions are met, the approximation performs poorly when estimating the
probability of a small range of counts
4.4 Negative Binomial Distribution

What is it?
The negative binomial distribution is a more general version of the geometric distribution: It
th th
describes the probability of observing the  success on the  trial.

Conditions
l Independence
l Outcome is either success or failure
l  is the same for each trial
l The last trial being a success

The Probability
th
The negative binomial distribution describes the probability of observing the  success on
th
the  trial, where all trials are independent:

n  
 the  th success on the  th trial k  p k   p n  k

4.5 Poisson Distribution

What does it do?


The Poisson distribution helps to estimate the number of events in a large population over a
unit of time.

Parameter
The rate — or the average number of occurrences in a mostly-fixed population per unit of
time — is the parameter in the Poisson distribution. It is the number of events we expect to
observe, and it is denoted by  or 

Formulas
   
 observe  events  


  

These formulas can be easily proven by expanding the first equation to the form of a Taylor
series.

You might also like