# Curve Fitting

Printable View

• Sep 14th 2010, 04:00 PM
Apprentice123
Curve Fitting
Determine the equation of a straight that best fit the points:

Code:

X|  0  |  2 |  4 |  6  |  8  |  10  | Y|1.12 |5.06 |9.09 |13.12 |17.14 |21.11 |
• Sep 14th 2010, 04:12 PM
mr fantastic
Quote:

Originally Posted by Apprentice123
Determine the equation of a straight that best fit the points:

Code:

X|  0  |  2 |  4 |  6  |  8  |  10  | Y|1.12 |5.06 |9.09 |13.12 |17.14 |21.11 |

What method are you expected to use? Linear regression by hand? Technology? etc. Please be more specific about what method you're expected to use.
• Sep 14th 2010, 04:39 PM
Apprentice123
Quote:

Originally Posted by mr fantastic
What method are you expected to use? Linear regression by hand? Technology? etc. Please be more specific about what method you're expected to use.

Least Squares method
I do not understand is how to find the equation of the line. What would be the polynomial
• Sep 14th 2010, 07:32 PM
mr fantastic
Quote:

Originally Posted by Apprentice123
Least Squares method
I do not understand is how to find the equation of the line. What would be the polynomial

Please answer the question I asked in my first reply: Are you expected to do it using technology (eg. calculator that has a linear regression program) or by hand?
• Sep 14th 2010, 07:45 PM
Apprentice123
Quote:

Originally Posted by mr fantastic
Please answer the question I asked in my first reply: Are you expected to do it using technology (eg. calculator that has a linear regression program) or by hand?

Ohh, sorry. Is by hand
• Sep 15th 2010, 03:23 AM
HallsofIvy
Then what formulas do you know? I doubt that you are expected to derive the "least squares" solution by from scratch.
• Sep 15th 2010, 05:37 AM
Apprentice123
Quote:

Originally Posted by HallsofIvy
Then what formulas do you know? I doubt that you are expected to derive the "least squares" solution by from scratch.

I know the method of least squares, where I have several meeting points and a polinimio.
What polynomial I need to find in this case?
• Sep 15th 2010, 07:50 AM
HallsofIvy
There are several different formulas to compute "least squares". What formula are you using?
• Sep 15th 2010, 02:09 PM
Apprentice123
Quote:

Originally Posted by HallsofIvy
There are several different formulas to compute "least squares". What formula are you using?

Is this:

$\displaystyle \sum_{k=1}^{m} (d_k)^2 = \sum_{k=1}^{m} [f(x_k) - \phi (x_k)]^2$

$\displaystyle \phi (x) = \alpha_1 g_1(x) + ... + \alpha_n g_n(x)$

$\displaystyle F(\alpha_1 , \alpha_2 , ..., \alpha_n ) = \sum_{k=1}^{m} (d_k)^2 = \sum_{k=1}^{m} [f(x_k) - \phi (x_k)]^2$