User's Manual

20070201
• Linear Regression ...
M
Se =
Σ
1
n – 2
i=1
n
(yi – (axi + b))
2
• Quadratic Regression ...
M
Se =
Σ
1
n – 3
i=1
n
(y i – (ax i + bx i + c))
2
2
• Cubic Regression ...
M
Se =
Σ
1
n – 4
i=1
n
(yi – (axi
3
+ bxi + cxi + d ))
2
2
• Quartic Regression ...
M
Se =
Σ
1
n – 5
i=1
n
(y i – (ax i
4
+ bx i
3
+ cx i + dx i + e))
2
2
• Logarithmic Regression ...
M
Se =
Σ
1
n – 2
i=1
n
(y i – (a + b ln x i ))
2
• Exponential Repression ...
M
Se =
Σ
1
n – 2
i=1
n
(ln y i – (ln a + bx i ))
2
• Power Regression ...
M
Se =
Σ
1
n – 2
i=1
n
(ln y i – (ln a + b ln x i ))
2
• Sin Regression ...
M
Se =
Σ
1
n – 2
i=1
n
(y i – (a sin (bx i + c) + d ))
2
• Logistic Regression ...
M
Se =
Σ
1
n – 2 1 + ae
-bx
i
C
i=1
n
y i –
2
u Estimated Value Calculation for Regression Graphs
The STAT mode also includes a Y-CAL function that uses regression to calculate the
estimated y -value for a particular x -value after graphing a paired-variable statistical
regression.
The following is the general procedure for using the Y-CAL function.
1. After drawing a regression graph, press !5 (G-SLV)1 (Y-CAL) to enter the graph
selection mode, and then press w .
If there are multiple graphs on the display, use f and c to select the graph you
want, and then press w .
• This causes an x -value input dialog box to appear.
6-4-4
Performing Statistical Calculations