D.4 Plotting

다음은 추정한 결과로 그림을 그린 결과이다.

#dev.new()
par(mfrow=c(1, 3))
plot(r1, main="Survival Probability", xlab="Time", ylab="Survival Probability")
plot(r1, type="cumhaz", main="Cumulative Hazard", xlab="Time", ylab="Cumulative Hazard")
plot(r1, type="hazard", main="Hazard", xlab="Time", ylab="Hazard Rate")
Survival and hazard esimation of the base model

Figure D.1: Survival and hazard esimation of the base model

D.4.1 Difference by Sex

성별에 따른 차이를 보고 싶으면 다음과 같이 할 수 있다.

r2 = flexsurvreg(Surv(TIME, DV) ~ SEX, data=d1, dist="gompertz") ; r2
Call:
flexsurvreg(formula = Surv(TIME, DV) ~ SEX, data = d1, dist = "gompertz")

Estimates: 
       data mean  est       L95%      U95%      se        exp(est)  L95%      U95%    
shape        NA    0.02868   0.01781   0.03956   0.00555        NA        NA        NA
rate         NA    0.00422   0.00204   0.00874   0.00157        NA        NA        NA
SEX     0.48000    0.35980  -0.21074   0.93033   0.29109   1.43304   0.80999   2.53535

N = 50,  Events: 48,  Censored: 2
Total time at risk: 2755
Log-likelihood = -229, df = 3
AIC = 464
#dev.new()
par(mfrow=c(1, 3))
plot(r2, main="Survival Probability", xlab="Time", ylab="Survival Probability")
plot(r2, type="cumhaz", main="Cumulative Hazard", xlab="Time", ylab="Cumulative Hazard")
plot(r2, type="hazard", main="Hazard", xlab="Time", ylab="Hazard Rate")
Survival and hazard of the covariate model

Figure D.2: Survival and hazard of the covariate model

만약 성별에 따라 별도로 그리고 싶다면 다음과 같이 할 수 있다.

#dev.new()
par(mfrow=c(1, 2))
plot(r2, newdata=d1[d1$SEX==0, ], main="SEX=0", xlab="Time", ylab="Survival Probability")
plot(r2, newdata=d1[d1$SEX==1, ], main="SEX=1", xlab="Time", ylab="Survival Probability")
Separate plot by sex

Figure D.3: Separate plot by sex