Casio FX6300G - Page 72

•The regression formula is >>=A + B*lnx. Enter the x data as the logarithm (In)
of x, and the y data inputs the same as that for linear regression.
•Estimated values x, and y based on the regression formula can be calculated
using the following formulas:
j>=A+B-Inx x=exp^ —
■Logarithmic regression
•The same operation as with linear regression can be used to obtain the regres
sion coefficient and for making corrections. To obtain the estimated value
O S *® ® ® ¡s used, and to obtain estimated value Jc, is
used.
Furthermore, Lx, Lx2 and Lxy are obtained by Elnx, £(lnx)2, and Elnx*y
respectively.
Operation Display
(SHÏFTliSdlÎBŒl
(Clears memory)
Q ïï)29 ® [ 3 1 .6 (dt|
3.36729583
G1Q50®Gl)23.5(dt) 3.912023005
Qn) 74 ® Q 38.0 63
4.304065093
Q31 0 3 ® (Z)46.4(dt)
4.634728988
Q«D118®[I)48.90>t)
4.770684624
(Constant term A)
(shift) ip Ira)
-1 1 1 .1 2 83 9 7 6
(Regression coefficient B)
S E S
34.02014749
(Correlation coefficient r)
m m m
0.9 940139464
(y when */= 80) Q n )80® ® (^ |
37.94879482
(x when yi- 73)
22 4.1541314
Example
xi
yi
29
1.6
50
23.5
74
38.0
103
46.4
118
48.9
The data in the above table
can be used to obtain the
terms of the regression for
mula and the correlation coeffi
cient. Based on the regression
formula, estimated value y can
be obtained for */= 80, and
estimated value x can be
obtained for yi = 73.
- 7 1 -