Chi Squared Lin Reg Flashcards
Why is it not appropriate to test hypothesis using the chi square goodness of fit test
You need counts for chi squared not averages
To find the appropriate p value of the chi square test statistic you use
Use Chi squared cdf
Plug in test statistic 8.3
Da grew of freedom 8
To find degree of freedom you use
Degree of freedom = #category - 1
All chi squared curves are skewed to the right?
True
Chi squared distribution are continuous distribution?
True
They go to Infiniti
The data must be from a reasonably random sample?
True
Chi squared statistic will be positive or negative depending on the way that you subtract?
False
All will be positive and always goes to the right
The chi squared test statistic is a measure of how different the observed counts are from the expected counts
True
To calculate expected counts for a two way table, one would multiply the table total by the row total and divide by the column total
False
( Row total x column total ) / table total
For chi square test and GOF the assumption are
Simple random sample
Expected counts are at least 5 or they are greater or equal to 5
Expected counts are (list)
If you have a matrix or a table you use what test
You use the chi squared test
What happens to the chi squared distribution (density curve) as degree of freedom increase
The chi squared density curve will shift to the right and becoming more like a normal distribution
In a chi squared test what are the
Ho:
Ha:
Ho: independent
Ha: dependent
In a chi squared GOF the
Ho:
Ha:
Are?
Ho: observed = expected
Ha: observed does not equal expected
Ho: same
Ha: different
If you have a list or a ratio you use what test?
Chi squared goodness of freedom test
The sum of the residuals is
Zero
If the residual plot is random the data is
Linear
If residual plot not random then it is
Not linear
The coefficient of determination is
R squared
Coefficient of determination in context is said like
% of the variation in (. ) can be explained by the LSRL
What is the correlation coefficient
R
Interpret the correlation coefficient by saying
There is a (strong, moderate, weak) (positive or negative) correlation between (blank)
Interpret slop by saying
For each additional unit of x there is an approx increase in y
For a linear regression t test what does it test?
It test whether slope is zero or not
Linear regression t test
Ho:
Ha:
Is what?
Ho: B=0
Ha: B does not equal zero
Where B is the true slope of the LSRL
When putting data into from the calculator for a linear regression t test you put the
X
Y in what
X goes in L1
Y goes in L2
What is the standard deviation or standard error
It is s
What is the slope standard deviation
Sb
Where is the y intercept and the slope?
The y intercept is always above the slope in the given info
Where is the slope standard deviation located in the info
It is located to the right of the slope
Strong positive linear relation ship is between
.8-1
Moderate positive linear relationship
.6-.8
Weak positive relation ship is between
> 0-.6
How to find t in linear regression problem
T= slope/standard deviation slope
Then you put it in tcdf