Curve Expert is a Windows application that allows the user to determine the best fit for XY data points using a variety of curve fitting models (over 30) including linear, quadratic, power, exponential, sinusoidal, logarithmic. Nonlinear regression models, polynomial interpolation and splines are also included.
The user may specify the model to be generated, or may ask the program to select a best fitting model; the best fit is determined by standard error and correlation coefficients. Custom regression models may be user defined.
The data, as well as the plotted curves, are given graphically to show which model fits best. Standard error and correlation coefficient are reported.
Type of Material:
Simulation, reference tool.
Recommended Uses:
This is more of a reference material, but a lesson that discusses different ways to fit a curve to data can easily incorporate Curve Expert.
Technical Requirements:
Windows operating system, PKUNZIP to unzip the compressed file. The reviewed application is the shareware version of Curve Expert. Individual licensed copies of the most recent version may be purchased for a nominal fee. Site licenses are also available, and 50% discounts are provided for educational institutions.
Identify Major Learning Goals:
This is more of a reference material, but a lesson that discusses different ways to fit a curve to data can easily incorporate Curve Expert.
Target Student Population:
Anyone with a need to fit a curve to data. The software is appropriate for a precalculus or elementary statistics student, as well as for more advanced mathematics applications.
Prerequisite Knowledge or Skills:
Functions, plotting data points, linear regression.
Content Quality
Rating:
Strengths:
A multitude of curve fitting and interpolation models can be brought to bear on a given data set.
Data is easily manipulated using the features of swapping, sorting, translating and scaling.
Curve Finder will automatically determine the best fitting model based on standard error and correlation coefficients.
Graphical output can be used to visually inspect the selected fit. Data about the fitted curves (integrals, derivatives, etc.) can be computed once the fit is complete.
This program has effectively combined all essential features in data analysis into one easy to use format.
Concerns:
In the analysis part for the fitted graphs integrals and arclengths sometimes are not computed correctly. Warnings are given if this occurs though. If the computation is possible after all, the reader could take the expression of the approximating curve and perform the computation with another tool.
Potential Effectiveness as a Teaching Tool
Rating:
Strengths:
This is a reference tool that because of its versatility can be the perfect support tool for classes that require data analysis and curve fitting.
The sheer number of curve fitting models can be used to stimulate discussion on various approaches to curve fitting. For example linear data with one point gone awry can come up fit best by a model other than linear. This could start a discussion on outliers as well as on the value of checking an approximation graphically.
The interpolation models can be used in a section on interpolation in a numerics class to quickly show how polynomial interpolation can oscillate widely on certain data sets, thus showing that going through all data points is not necessarily a goodness-of-fit indicator.
Concerns:
None.
Ease of Use for Both Students and Faculty
Rating:
Strengths:
Spreadsheet type interface.
Easy import of data from files and save of data to files.
Printing includes full-page graphs. The user can adjust the colors of the data points and curve,
change the symbols of the plots, and change the style of the curve.
Fitting models are well-organized, with categories such as exponential and sigmoidal offering several subcategories.
Data tables, as well as the fitting curve, are displayed graphically.
Once a model equation is selected, its coefficients, covariance matrix, residuals, and error information are available by right-clicking the mouse and selecting ?Information?.
Curve finder automates the hard work of finding the right model. The user can specify a subset of the available models, and Curve Expert will create each model with the best fit listed first.
Well-written documentation.
Concerns:
The ability to superimpose various curves on the same plot would be useful. It may be present, but neither reviewer was able to find it.
Creative Commons:
Search by ISBN?
It looks like you have entered an ISBN number. Would you like to search using what you have
entered as an ISBN number?
Searching for Members?
You entered an email address. Would you like to search for members? Click Yes to continue. If no, materials will be displayed first. You can refine your search with the options on the left of the results page.
Searching for Members?
You entered an email address. Would you like to search for members? Click Yes to continue. If no, materials will be displayed first. You can refine your search with the options on the left of the results page.
