edited tags

Source Link

edited Dec 6, 2012 at 21:10

333.5k
63
792
1.3k

p<-glm(GRADE~GPA+TUCE+PSI,family="binomial"(link="probit")) summary(p)

p<-glm(GRADE~GPA+TUCE+PSI,family="binomial"(link="probit")); summary(p)

is the probit model

Coefficients: Estimate Std. Error z value Pr(>|z|)
(Intercept) -7.45231 2.57152 -2.898 0.00376 **

GPA 1.62581 0.68973 2.357 0.01841 *

TUCE 0.05173 0.08119 0.637 0.52406

PSI 1.42633 0.58695 2.430 0.01510 *

Coefficients:
              Estimate    Std. Error  z value   Pr(>|z|)   
(Intercept)   -7.45231    2.57152     -2.898    0.00376 **
GPA            1.62581    0.68973      2.357    0.01841 * 
TUCE           0.05173    0.08119      0.637    0.52406  
PSI            1.42633    0.58695      2.430    0.01510 *

If we have mean of TUCE anand PSI=1, what is the marginal effect of GPA on Pr(GRADE=1)?

pnorm(-7.45231+1.62581mean(GPA)+0.05173mean(TUCE)+1.42633*1)*1.62581

dnorm(-7.45231+1.62581mean(GPA)+0.05173mean(TUCE)+1.42633*1)*1.62581

pnorm(-7.45231+1.62581*mean(GPA)+0.05173*mean(TUCE)+1.42633*1)*1.62581  
dnorm(-7.45231+1.62581*mean(GPA)+0.05173*mean(TUCE)+1.42633*1)*1.62581

I'm wondering which one of pnormpnorm and dnormdnorm is correct to use if you want the marginal effect.

p<-glm(GRADE~GPA+TUCE+PSI,family="binomial"(link="probit")) summary(p)

is the probit model

Coefficients: Estimate Std. Error z value Pr(>|z|)
(Intercept) -7.45231 2.57152 -2.898 0.00376 **

GPA 1.62581 0.68973 2.357 0.01841 *

TUCE 0.05173 0.08119 0.637 0.52406

PSI 1.42633 0.58695 2.430 0.01510 *

If we have mean of TUCE an PSI=1, what is the marginal effect of GPA on Pr(GRADE=1)?

pnorm(-7.45231+1.62581mean(GPA)+0.05173mean(TUCE)+1.42633*1)*1.62581

dnorm(-7.45231+1.62581mean(GPA)+0.05173mean(TUCE)+1.42633*1)*1.62581

I'm wondering which one of pnorm and dnorm is correct to use if you want the marginal effect.

p<-glm(GRADE~GPA+TUCE+PSI,family="binomial"(link="probit")); summary(p)

is the probit model

Coefficients:
              Estimate    Std. Error  z value   Pr(>|z|)   
(Intercept)   -7.45231    2.57152     -2.898    0.00376 **
GPA            1.62581    0.68973      2.357    0.01841 * 
TUCE           0.05173    0.08119      0.637    0.52406  
PSI            1.42633    0.58695      2.430    0.01510 *

If we have mean of TUCE and PSI=1, what is the marginal effect of GPA on Pr(GRADE=1)?

pnorm(-7.45231+1.62581*mean(GPA)+0.05173*mean(TUCE)+1.42633*1)*1.62581  
dnorm(-7.45231+1.62581*mean(GPA)+0.05173*mean(TUCE)+1.42633*1)*1.62581

I'm wondering which one of pnorm and dnorm is correct to use if you want the marginal effect.

probit marginal-distribution

added 484 characters in body

Source Link

edited Dec 6, 2012 at 19:54

rlost

11
1
4

p<-glm(GRADE~GPA+TUCE+PSI,family="binomial"(link="probit")) summary(p)

is the probit model

Coefficients: Estimate Std. Error z value Pr(>|z|)
(Intercept) -7.2+145231 2.53+0.122+157152 -2.7*1898 0.00376 **

GPA 1.62581 0.68973 2.357 0.01841 *

TUCE 0.05173 0.08119 0.637 0.52406

PSI 1.42633 0.58695 2.430 0.01510 *

If we have mean of TUCE an PSI=1, what is a probit model obtained using glmthe marginal effect of GPA on Pr(GRADE=1)?

pnorm(-7.2+145231+1.5625813+0mean(GPA)+0.10517322+1mean(TUCE)+1.7*142633*1)*1.562581

dnorm(-7.2+1.5*+045231+1.16258122+1mean(GPA)+0.7051731mean(TUCE)+1.42633*1)*1.562581

I'm wondering which one of pnorm and dnorm is correct to use if you want the marginal effect.

Source Link

asked Dec 6, 2012 at 19:12

rlost

11
1
4

Probit model marginal effects

-7.2+1.53+0.122+1.7*1 is a probit model obtained using glm

pnorm(-7.2+1.53+0.122+1.7*1)*1.5

dnorm(-7.2+1.5*+0.122+1.71)*1.5

I'm wondering which one of pnorm and dnorm is correct to use if you want the marginal effect.

marginal-distribution

Stack Exchange Network

Return to Question

Probit model marginal effects