Least-Squares Polynomials
- Because polynomials can be readily manipulated, fitting such functions to data that do not plot linearly is common.
- It will turn out that the normal equations are linear for this situation, which is an added advantage.
- as the degree of the polynomial and N as the number of data pairs. If , the polynomial passes exactly through each point and the methods discussed earlier apply, so we will always have in the following. We assume the functional relationship
|
(5.5) |
with errors defined by
We again use to represent the observed or experimental value corresponding to , with free of error. We minimize the sum of squares;
At the minimum, all the partial derivatives
vanish. Writing the equations for these gives equations:
Dividing each by and rearranging gives the normal equations to be solved simultaneously:
|
(5.6) |
Putting these equations in matrix form shows the coefficient matrix;
|
|
|
(5.7) |
All the summatins in Eqs. 5.6 and 5.7 run from 1 to . We will let B stand for the coefficient matrix.
- Equation 5.7 represents a linear system. However, you need to know that this system is ill-conditioned and round-off errors can distort the solution: the 's of Eq. 5.5. Up to degree-3 or -4, the problem is not too great. Special methods that use orthogonal polynomials are a remedy. Degrees higher than 4 are used very infrequently. It is often better to fit a series of lower-degree polynomials to subsets of the data.
- Matrix of Eq. 5.7 is called the normal matrix for the least-squares problem. There is another matrix that corresponds to this, called the design matrix. It is of the form;
is just the coefficient matrix of Eq. 5.7. It is easy to see that , where is the column vector of -values, gives the right-hand side of Eq. 5.7. We can rewrite Eq. 5.7 in matrix form, as
- It is illustrated the use of Eqs. 5.6 to fit a quadratic to the data of Table 5.4. Figure 5.7 shows a plot of the data. The data are actually a perturbation of the relation
.
Table 5.4:
Data to illustrate curve fitting.
|
Figure 5.7:
Figure for the data to illustrate curve fitting.
|
To set up the normal equations, we need the sums tabulated in Table 5.4. The equations to be solved are:
The result is
,
,
, so the least- squares method gives
which we compare to
. Errors in the data cause the equations to differ.
2004-12-28