Curve Fitting And Forecasting - HP 17bII Owner's Manual

Curve Fitting and Forecasting

Curve fitting is a statistical method for finding a relationship between
two variables, x and y. Based on this relationship, you can estimate new
values of y based on a given x-value, and vice-versa. Each SUM list
holds the numbers (data values) for one variable. You can select one of
four curve-fitting models:
y
Logarithmic Curve Fit
y
*
The exponential, logarithmic, and power models are calculated using
transformations that allow the data to be fitted by standard linear regression.
The equations for these transformations appear in appendix B. The
logarithmic model requires positive x-values; the exponential model requires
positive y-values; and the power curve requires positive x- and y-values.
File name : English-M02-1-040308(Print).doc
*
Linear Curve Fit
x
x
10: Running Total and Statistics 133
Print data : 2004/3/9
Exponential Curve Fit
y
Mx
Power Curve Fit
y
M
x
x