cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
%3CLINGO-SUB%20id%3D%22lingo-sub-574689%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3EBox%20Tidwell%20pour%20la%20r%C3%A9gression%20logistique%3C%2FLINGO-SUB%3E%3CLINGO-BODY%20id%3D%22lingo-body-574689%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3E%3CP%3ESalut%20la%20communaut%C3%A9%20JMP%2C%3C%2FP%3E%3CP%3EL'une%20des%20hypoth%C3%A8ses%20de%20la%20r%C3%A9gression%20logistique%20est%20la%20lin%C3%A9arit%C3%A9%20des%20variables%20et%20le%20log%20des%20cotes.%20Cependant%2C%20je%20n'ai%20toujours%20pas%20compris%20comment%20tester%20cela%20dans%20JMP.%20Le%20test%20sp%C3%A9cifique%20que%20j'ai%20vu%20ailleurs%20est%20le%20test%20Box%20Tidwell.%20Existe-t-il%20un%20moyen%20d'obtenir%20ce%20test%20sur%20JMP%26nbsp%3B%3F%20A%20d%C3%A9faut%2C%20existe-t-il%20un%20autre%20moyen%20de%20tester%20la%20lin%C3%A9arit%C3%A9%20d'un%20mod%C3%A8le%20de%20r%C3%A9gression%20logistique%26nbsp%3B%3F%3C%2FP%3E%3CP%3EMerci!%3C%2FP%3E%3C%2FLINGO-BODY%3E%3CLINGO-LABS%20id%3D%22lingo-labs-574689%22%20slang%3D%22en-US%22%20mode%3D%22NONE%22%3E%3CLINGO-LABEL%3EMod%C3%A9lisation%20statistique%20avanc%C3%A9e%3C%2FLINGO-LABEL%3E%3CLINGO-LABEL%3EAnalyse%20et%20mod%C3%A9lisation%20des%20donn%C3%A9es%20de%20base%3C%2FLINGO-LABEL%3E%3C%2FLINGO-LABS%3E
Choose Language Hide Translation Bar
View Original Published Thread

0 Kudos

Box Tidwell pour la régression logistique

Salut la communauté JMP,

L'une des hypothèses de la régression logistique est la linéarité des variables et le log des cotes. Cependant, je n'ai toujours pas compris comment tester cela dans JMP. Le test spécifique que j'ai vu ailleurs est le test Box Tidwell. Existe-t-il un moyen d'obtenir ce test sur JMP ? A défaut, existe-t-il un autre moyen de tester la linéarité d'un modèle de régression logistique ?

Merci!

Ce message écrit à l'origine en English (US) a été traduit pour votre commodité. Lorsque vous répondez, il sera également traduit en English (US).

3 Comments
Sarah-Sylvestre
Staff
Status changed to: Acknowledged

Hi @aande2, thank you for your suggestion! We have captured your request and will take it under consideration.

SamGardner
Level VII
Status changed to: Investigating
 
SamGardner
Level VII
Status changed to: Not Planned For Now

Hello @aande2 after some research we have decided to not pursue this.  You can do this in JMP manually by adding in the interactions or each continuous term with its log transform. That is, if Logit(Y) = aX, then fit the model Logit(Y) = a X + b X*Ln(X).  If the parameter estimate for the X*Ln(X) effect is significant, then Logit(Y) is deemed not linearly dependent on X. 

 

Run this script to see how to do this:

 

dt=Open( "$SAMPLE_DATA/Ingots.jmp" );

// Logistic regression model
dt << Fit Model(
	Freq( :count ),
	Y( :ready ),
	Effects( :heat, :soak ),
	Personality( "Nominal Logistic" ),
	Run( Likelihood Ratio Tests( 1 ), Wald Tests( 0 ) )
);

// model with Box Tidwell effects
dt <<Fit Model(
	Y( :ready ),
	Effects( :heat, :soak, :heat * :"Log[heat]"n, :soak * :"Log[soak]"n ),
	Target Level( "Ready" ),
	Personality( "Nominal Logistic" ),
	Run( Likelihood Ratio Tests( 1 ), Wald Tests( 0 ) )
);