590 Chapter 10. Software for Data Analysis
Figure 10.1.16: Plotting a time series and performing FFT in Origin.limitation that the data must be regularly spaced and must have been sampled at
or above the Nyquist frequency (see chapter on data analysis and statistics). For
irregularly spaced data one can useLombmethod instead of Fourier transform.
Let us now move to another example. What we now want is to fit a nonlinear
function to some data. We follow the same steps as outlined in the previous example
to import the data and to plot it. Then from theToolspull down menu we select
Nonlinear fit wizardmethod. A pop up window appears where one could choose the
fitting function or define a new one. Origin has a very large number of built in
functions and one rarely needs to define another one. For our example, we chose
to fit a 5th order polynomial to the data. The result is shown in Fig.10.1.17. The
statistics related to the fit are also displayed along with the plot. The fitted curve
itself is superimposed on the actual data plot, giving the user a qualitative way to
judge the fit. Another very useful feature of fitting wizard is that it also shows a
visual of the selected fitting function, which gives the user some idea of whether the
function would be a good fit to the data or not.
All of the fitting functions available in the fitting wizard are also available in
Origin C. Hence, if one decides to write a code for data analysis, all these methods
can be called by using built in Origin C functions. This route is generally chosen
for analysis of large datasets that have to be analyzed repeatedly through the same
method. In such a case one is better off writing a code once and running it each
time instead of invoking the menu driven functions time and again.