Multiple Regression in Excel

preview_player
Показать описание
Explains how to use multiple regression in Excel 2010. Made by faculty at the University of Colorado Boulder Department of Chemical and Biological Engineering.
Check out our Kinetics Review playlist here:
Рекомендации по теме
Комментарии
Автор

A better way to solve this problem is by nonlinear regression on the original form of the equation; this can be set up using guessed initial coefficient values, computing the sum of squared residuals or SSR (residual = model-data values; on the spreadsheet, sum the squares of all residuals to get the SSR), then use Solver to minimize the SSR by changing the coefficient values. Those are the true best fit values for the original equation. Try both methods and graph them to convince yourself.

skdentel
Автор

The multiple regression demonstration is fine, but the conversion of a power equation to a linearized equation is a poor way to determine best fit coefficients. The regression's best fit values may be optimal for the logarithmic form of the equation, but they are NOT likely to be the best fits for the original form of the equation.

skdentel