# Thread: Problems in Cox proportional hazards model, Survival analysis.

1. ## Problems in Cox proportional hazards model, Survival analysis.

Hello everyone, I got trouble with some problems as following

Suppose a Cox proportional hazards model was fit with 2 covariates, smoking status X_1 and age X_2. The smoking status X_1 was coded X_1=4 for smokers and X_1=0 for non-smokers. The age X_2 was in years on a continuous scale.

(a) If X_1 is now recorded X_1=2 for smokers and X_1=0 for non-smokers, how will the relative risk change? How will the regression coefficient beta_1 change?

(b) It was found that age was under-reported by 3 years for everyone. A second analysis was performed by adding 3 years to everyone's age. How will the regression coefficient beta_2 change?

For (a), I thought that since the weight for smokers is decreased, beta_1 should be decreased too. And for (b), since all ages are shifted, I thought there should be no difference for beta_2.

But I wonder if what I thought is right. Can any one help me? Thank you!

2. ## Re: Problems in Cox proportional hazards model, Survival analysis.

Hi, can anybody help me with the problem like what is meant by this formula, P = 1- cdf (chi2, df, logLR)?

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts