Functions > Data Analysis > Curve Fitting > About Curve Fitting Functions
When you perform curve fitting, you fit a single function, in a least-squares sense, through all data points. This method contrast with interpolation, where piece-wise functions are fitted through adjacent data points.
To further analyze your data, or determine the suitability of the chosen regression, you may wish to apply other statistics functions for data analysis.
Linear and Median-Median Regression
line, slope, intercept, stderr—Least-squares linear regression for data, and the standard error associated with linear regression
medfit—Median-median line regression for data
Polynomial and Rational Function Regression
loess—Localized polynomial regression
rationalfit, rationalfitnp—Least-squares rational function regression
polyfit, polyfitc, polyfitstat—Multivariate polynomial regression
Nonlinear Regression
genfit—Least-squares nonlinear regression for arbitrary fit functions
expfit—Least-squares exponential regression
lnfit, logfit—Least-squares logarithmic regression
lgsfit—Least-squares logistic curve regression
pwrfit—Least-squares power curve regression
sinfit—Least-squares sinusoidal regression
If you want to include additional information about the data or the parameters for any of the above fits, such as standard deviation in the data, bounds on the parameters, or constraint functions, you should use the LeastSquaresFit function to do the calculation in a more detailed way.
Other Functions
linfit—Least-squares regression for an arbitrary linear combination of functions
multidfit—General multivariate fit