Curve-Fitting Procedure CFP

The algorithm is essentially a multipass data scanning method Scan the data array and calculate the delta Y between adjacent pairs of points for the entire spectrum and take the average. This is considered the noise value. Scan the data array again, then do the following Look for data points where the delta Y is greater than the noise value and the delta Y value is positive. This is the start of a peak. Continue scanning and look for a local maximum point (Y values of the data points 

FIGURE 2.9 (1) Nonprocedured and (2) deconvoluted IR spectra of the liquid crystal MLC-6815 within the 1475 and 1430 cnr IR region (y= 3.8 and increment 0.1). The procedure is performed by GRAMS AI/7.01, an IR spectroscopic software package. 

As always, no method is perfect and, unfortunately, since there are a number of answers to a nonlinear problem, on many occasions the fit will end up in a local minimum, which may not be the best possible solution. Many fit routines will then continue to iterate using solutions that are significantly different from the minimum, trying to find another better solution. If one is not found, then the minimized solution is considered best. 

Other problems occur if the user input values are significantly far from the real answer. This may end in a solution that is stuck in a local minimum that is determined to be the best fit. In extreme cases the fit may even walk away to a ridiculous solution. The only way to recover from this problem is to reenter the starting fit values. Overfltting is another common problem. Better fits can always be obtained with more input variables (given enough variables you can fit anything). Thus it is important that the input parameters reflect the proper number of variables based on the physical measurement. 

---