BSNS102 4
Card Set Information
Author:
Anonymous
ID:
90022
Filename:
BSNS102 4
Updated:
2011-06-10 06:38:45
Tags:
Statistics
Folders:
Description:
Chi Square tests, Regression, Correlation
Show Answers:
Goodness of fit test:
Compares an observed set of frequencies with an expected set of frequencies
Test of independance:
Checks to see if independance holds in the population
X
^{2 }
test Conditions:
Random Sample
Sample is large
Goodness of fit test
Conditions:
Fixed number of trials
Constant probability of success
Independant trials
X
^{2 }
test of independance
Conditions:
Random Sample
Large sample
Regression Analysis is used primarily to model :
Causality and provide prediction (forecasting).
To predict the values of a dependant variable based on values of at least one independant variable.
Explain the effect of the independant variables on the dependant variable.
Types of Relationships
Positive Linear Relationship
Negative Linear Relationship
Relationship Not Linear
No Relationship
Regression Conditions:
Errors
: (distance between a value of y and the population regression line) are:
Independant of the independant variable (x)
Do not systematically vary with the dependant variable (y)
Are independant of time
Normality
:
The errors are normally distributed atound the regression line
Homoscedasticity
(constant variance):
For each x value, the variable of y around the regression line is the same
Correlation measures:
the degree of linear association between 2 numerical values
Covariance measures:
how one numerical variabl linearly co-varies with another numerical variables.
Strength of Correlation:
Very Strong
: r > 0.90
Strong
: 0.75 < r < 0.90
Moderate
: 0.4 < r < 0.75
Weak
: r < 0.4
Virtually no correlation
: r is close to zero
Z confidence levels:
90% = 1.645
95% = 1.96
99% = 2.576