OriginLab Corporation - Data Analysis and Graphing Software - 2D graphs, 3D graphs, Contour Plots, Statistical Charts, Data Exploration, Statistics, Curve Fitting, Signal Processing, and Peak Analysis                           
 

3.22 FAQ-248 How do I interpret polynomial fitting results?

Last Update: 11/9/2018

Origin provides two ways to fit a polynomial to your data:

  • The Polynomial Fit tool, accessible from the Analysis: Fitting: Polynomial Fit... menu item.
  • The Nonlinear Curve Fitter tool, accessible from the Analysis: Fitting: Nonlinear Curve Fit... menu item. In this tool you can then select the Polynomial function category and a suitable function, or define your own fitting function.

When fitting with a polynomial equation, the objective typically is to obtain a fit curve that follows the data as closely as possible. When too many higher order terms are involved, it is possible that the fit curve may swing above and below data points. If this occurs, try reducing the number of terms. If the fit curve is farther away from most data points, try increasing the number of terms.

If you plan to use the coefficients reported by the fit elsewhere in Origin or in outside applications, please note that the number of significant digits used when copying the coefficient values may greatly alter the subsequent computation.

You will therefore need to do the following, to copy the coefficient values with full precision:

  1. Go to the fit report sheet
  2. In the Parameters table, drag and select the Value cells
  3. With the selection active, use the Edit->Copy->Copy (full precision) menu item

Now you can paste these full-precision values elsewhere.

Please refer to this FAQ for further information on copying values with full precision in other contexts within Origin.

If you simply want to plot a curve that closely resembles your data and you do not need to interpret or use the fit coefficients, you may try using the interpolation tools available under the Analysis: Mathematics menu, instead of performing a polynomial fit. These tools offer linear, spline and b-spline interpolation with variable number of points for the output curve.


Keywords:high, order, terms, wiggle, swing, residue, bad

 

© OriginLab Corporation. All rights reserved.