You will always benefit if you can employ your knowledge of a system as part of the modeling process.
Only you know your data, as the scientist, engineer, orĪnalyst. Your eye and brain are splendid at things like pattern recognition. In fact, I'll suggest that you should plot everything. It is always a good idea to plot your data. An exponential is a good place to start, a simple curve shape that is easy to fit. Plus, I want to assure an understanding of polynomials, since many of the tools for interpolation So I'm starting out with some discussion about Produced by a regression model, calling both of these things interpolation. Many people mistake the ideas of interpolation with the approximation Along the way I'll try to give some pointers on curve fitting, interpolation,Ī valid question for some to ask is why start out with a discussion about polynomial regression, when we really wanted to talk about interpolation. I.e., find a curve that passes through your data. Only talk about problems with one independent variable.) In these coming blogs, I'll try to show some ways to do exactly this, I'll assume you have some data points through which you wish to pass a curve, interpolating your data.