Lesson 32: regression analysis of paired data, Best fit – HP 48G User Manual
Page 93
Attention! The text in this document has been recognized automatically. To view the original document, you can use the "Original mode".
Lesson 32: Regression Analysis of Paired Data
The HP 48
can
fit any one of four
statistical models
to a
set
of data
pairs. Once again the data must be in the
S D A T
matrix.
Example:
Using
the
same data
as the previous
example,
calculate
the correlation and covariance between the C'PI and PPI
variables using
the
best-fitting regression model
(of the
four available). Then predict the PPI value for a CPI
value of 8.5 using the calculated model.
S t e p 1 :
Open the STAT application and select Fbi. t D-d
CÖ
(D®
............... I
FIT DfiTfli
IDfiT:
___________
K-CDL: 1 Y-CDL: 2
MODEL: Linear Fit
ENTER iTDTISTICflL JflTfl_____________
irraiBHnnTHiMiiBiKai
S t e p 2 :
Check your regression
options.
The correct
data
is
in SDAT
already. The two variables tliat you wish to compare,
CPI and PPI. are located in columns 1 and 2 of SDAT,
respectively, so make sure that iv-COL» contains 1 and
Y.COLI: contains 2. Change the regression model type to
FIT DflTflä
EDflT: [[
9.1 9.2 ]
K-CDL: 1 Y-CDL:
2
MODEL:
Best Fit
CHOOSE STATISTICAL MODEL_______
Calculus, Statistics and Advanced Mathematics 7-9