HP (Hewlett-Packard) 50g ユーザーズマニュアル

Page 18-62

Because we will be using the same x-y data for fitting polynomials of different 
orders, it is advisable to save the lists of data values x and y into variables xx 
and yy, respectively.  This way,  we will not have to type them all over again in 
each application of the program POLY.  Thus, proceed as follows:

{ 2.3 3.2 4.5 1.65 9.32 1.18 6.24 3.45 9.89 1.22 } ` ‘xx’ K
{179.72 562.30 1969.11 65.87 31220.89 32.81 6731.48 737.41 39248.46 
33.45} ` ‘yy’ K

To fit the data to polynomials use the following:

@@xx@@ @@yy@@  2  @POLY, Result: [4527.73  -3958.52 742.23]
i.e.,                          y = 4527.73-3958.52x+742.23x

@@xx@@ @@yy@@  3  @POLY, Result: [ –998.05 1303.21  -505.27 79.23]
i.e.,                   y = -998.05+1303.21x-505.27x

@@xx@@ @@yy@@  4  @POLY, Result: [20.92 –2.61 –1.52 6.05 3.51 ]
i.e.,                      y = 20.92-2.61x-1.52x

@@xx@@ @@yy@@  5  @POLY, Result: [19.08 0.18 –2.94 6.36 3.48 0.00 ]
i.e.,              y = 19.08+0.18x-2.94x

@@xx@@ @@yy@@  6  @POLY, Result: [-16.73 67.17 –48.69 21.11 1.07 0.19 0.00]
i.e.,     y = -16.72+67.17x-48.69x

As you can see from the results above, you can fit any polynomial to a set of 
data.  The question arises, which is the best fitting for the data?  To help one 
decide on the best fitting we can use several criteria:

x

y

2.30

179.72

3.20

562.30

4.50

1969.11

1.65

65.87

9.32

31220.89

1.18

32.81

6.24

6731.48

3.45

737.41

9.89

39248.46

1.22

33.45