Power Curve

By Kardi Teknomo, PhD.

<Previous | Next | Contents>

Assuming that our model is power curve , we can take logarithm to both sides of the equation.

To compute manually, we take logarithm to both and . Thus, we have model transformation and .

Example

Data:

Then we plot log x versus log y and if the data is fit into a power curve, we will obtain a linear model in the double log plot. For our data we find with R-squared of 0.9999. Thus, the data fits into power curve. The parameters of the power curve can be obtained from the linear model using parameter transformation and . In this case, we have or . Thus, the regression line is with the same R-squared of 0.9999

<Previous | Next | Contents>

See Also:Regression tutorial, Power Rules, Logarithm Rules, Kernel Regression

Rate this tutorial or give your comments about this tutorial

This tutorial is copyrighted.

Preferable reference for this tutorial is

Teknomo, Kardi. Non-Linear Transformation for Regression. http:\\people.revoledu.com\kardi\ tutorial\Regression\NonLinear\