Effective Objects

Specifications:

• Fits Nonlinear and linear equations to raw data.

• EoFit is provided as a Win32 DLL with functions that can be used in programming languages that can call native Windows functions
• Example in Excel, VB 6, VB.NET, VC++, and C#.
• Clear documentation with plenty of examples and a tutorial to get started.
• Accepts equations in text strings with its built-in equation parser.
• Accepts equations in a user defined function.
• Allows bound limits to be put on parameters so that the results stay within a known range.
• Works with large data sets limited by the memory of the PC.
• Provides statistical information about results including stddev, residuals, residual stddev, etc.
• Tested with NIST standard nonlinear data sets.

What is least-squares fitting?

Least-square fitting finds the parameters of an equation that minimize the sum of the squares of the error between the fitted equation and the data. EoFit uses an adaptive optimization method to find parameters to the equation that minimizes this sum of the squares. For example, an equation such as y = a + b * x^c, can be fitted to the following data set. X=0,1,2,3 and y=4, 7, 16, 31. The equation and x and y data are passed into EoFit to produce the fitted parameters of a=4, b=3, and c=2. Much more complex problems are possible.

Will EoFit provide a fit for any equation?

EoFit is excellent for a wide class of equations that come up in normal applications. It uses an intelligent adaptive algorithm, unlike many solvers out there that are not adaptive. You can be assured EoFit provides the best modern nonlinear fitting available. The equation should be smoothly curving without discontinuities in the region to fit. EoFit allows you to select initial guesses and specify a limiting range for parameters to help narrow in on the solution for difficult problems.

What is a nonlinear equation?

A nonlinear equation is one that is nonlinear in the parameters to be fitted. Y = a + b*x + c*x^2 is linear in the parameters and nonlinear in the variable x. It is a linear equation in the parameters. However, Y = a + b*x^c is nonlinear in the parameter c so it is nonlinear. EoFit can solve both types of problem.

Least-square fitting finds the parameters of an equation that minimize the sum of the squares of the error between the fitted equation and the data. Visually this results in a smooth line running through the noisy points of data. Using proven methods, EoFit finds parameters to the equation that have the lowest least-squares error.

y = b1 * (b2+x)^(-1/b3)	&nbsp	y = c1 + c2 x + c3 x^2
y = c1 + c2 sin(x)	&nbsp	y = c1 + c2 sine(x) / x

What types of data can benefit from least-square fitting?

- Data acquisition readings from sensors including temperature, pressure, weight, strain, etc.

- Scientific data

- GPS position readings

- Weather data

- Sales activity data

- Manufacturing yield data

- Many other fields of data

This product includes a DLL library, pdf manual, and C++, VB, VB.NET, C# .NET and Excel examples to get your application running quickly. It works in any Windows 32-bit operating system.

EoFit includes a single station license to use in your application and for development. See the license agreement. A license to redistribute EoFit with your application is available for an additional $200. Please contact Effective Objects by phone or email for this option.

Compatibility:

Support
Support for installation and general questions is provided free. Technical questions regarding fitting specific functions is provided under our engineering support. Please purchase Engineering Development Support Credits in advance for this type of support. The demo above is highly recommended to evaluate this product.