6-5-16 Tests

u LinearReg t Test

LinearReg t Test treats paired-variable data sets as (x, y) pairs, and uses the method of least squares to determine the most appropriate a, b coefficients of the data for the regression formula y = a + bx. It also determines the correlation coefficient and t value, and calculates the extent of the relationship between x and y.

 

n

 

 

 

 

 

 

Σ( x o)( y p)

 

 

 

 

 

 

 

 

n – 2

b =

i=1

 

a = p bo

t = r

 

 

 

 

 

n

 

1 – r

2

 

 

 

 

Σ(x o)2

 

 

 

i=1

Perform the following key operations from the statistical data list.

3(TEST)

2(t)

3(REG)

a: intercept

b: slope of the line

n: size of sample (n > 3) r : correlation coefficient r2 : coefficient of

determination

The following shows the meaning of each item in the case of list data specification.

β & ρ ............................

p-value test conditions (“G 0” specifies two-tail test, “< 0”

 

specifies lower one-tail test, “> 0” specifies upper one-tail

 

test.)

XList .............................

list for x-axis data (List 1 to 26)

YList .............................

list for y-axis data (List 1 to 26)

Freq..............................

frequency (1 or List 1 to 26)

Save Res......................

list for storage of calculation results (None or List 1 to 26)

Execute ........................

executes a calculation

After setting all the parameters, use c to move the highlighting to “Execute” and then press the function key shown below to perform the calculation.

1(CALC) ... Performs the calculation.

# You cannot draw a graph for LinearReg t Test.

20070201