Comments for Other Types of Regression

For details about the calculation formula of the command included in each regression type, refer to the indicated calculation formulas

( Appendix <#069> to <#073>).

Statistical Calculation

Model Equation

Calculation

Type

Formula

 

 

 

Logarithmic Regression

y = A + BlnX

<#069>

(ln X)

 

 

 

 

 

 

 

 

 

e Exponential Regression

y = AeBX

<#070>

(e^X)

 

 

 

 

ab Exponential Regression

y = ABX

<#071>

(A•B^X)

 

 

 

 

Power Regression (A•X^B)

y = AXB

<#072>

 

 

 

 

 

Inverse Regression (1/X)

y = A +

B

 

<#073>

X

 

 

 

Appendix Comparison of Regression Curves

The following example uses the data input in Example <#061>. <#074> Compare the correlation coefficient for logarithmic, e

exponential, ab exponential, power, and inverse regression.

(FREQ: OFF)

Appendix Other Types of Regression Calculation: <#075> to

<#079>

Command Usage Tips

The commands included in the Reg sub-menu can take a long time to execute in logarithmic, e exponential, ab exponential, or power regression calculation when there are a large number of data samples.

Base-nCalculations

(BASE-N)

The BASE-N Mode lets you perform arithmetic calculations, negative value calculations, and logical operations with binary, octal, decimal, and hexadecimal values.

All calculations in this section are performed in the BASE-N Mode (N4).

E-52