Non Linear Regression

Nonlinear regression is a form of regression analysis in which observational data are modeled by a function which is a nonlinear combination of the model parameters and depends on one or more independent variables. The data are fitted by a method of successive approximations.

The Quant Express Non Linear Regression class determines the values of parameters for any equation, whose form you specify, that cause the equation to best fit a set of data values. It can handle linear, polynomial, exponential, logistic, periodic, and general nonlinear functions. Unlike many "nonlinear" regression programs that can only handle a limited set of function forms, our library can handle essentially any function whose form you can specify algebraically.

In this example, the random data are generated from the function y(x) with a0 = 2, a1 = 3 and a2 = -4.
 

y(x) = a0 + cos(a1 + x) + sin(a2 * x) (unconstrained) y(x) = a0 + log(a1 + x) + sin(a2 * x) (constrained)

NB: Multiple dimensions in X are supported too.
	Create Random Series and Fit the Data
X Y Estimated Y (Fitted) 1.000 1.869 1.907 2.000 1.709 2.489 3.000 3.141 2.766 4.000 3.332 2.438 5.000 0.375 1.745 6.000 1.738 1.300 7.000 0.770 1.546 8.000 1.130 2.319 9.000 4.913 2.946 10.000 2.331 2.833 11.000 2.197 2.024 12.000 1.943 1.213 13.000 -0.204 1.147 14.000 2.561 1.941 15.000 2.136 2.921 16.000 2.399 3.201 17.000 3.901 2.482 18.000 1.049 1.363 19.000 0.883 0.842 20.000 2.960 1.428 21.000 2.077 2.643 22.000 3.482 3.413 23.000 3.191 3.017 24.000 1.458 1.763 25.000 1.133 0.749 Loading…	Loading…