Dear all,
I'm running a Cox model using -stcox-. My main aim is to examine the effect of an independent continuous variable on the hazard to commit crime.
In the regression, the coefficient of this variable is positive and significant. However, I would like to "illustrate" the results and show what is the "predicted hazard" to commit crime, based on values of the independent variable. After linear regression, for example, I would use -margins- and get the predicted values of the dependent variable. But -margins- after -stcox- seems to give me something different.
So, my question is whether there is a good way to get predicted hazard to commit crime, i.e., to know what is the "risk" to commit crime during the follow-up period, based on values of my independent variable. Maybe using -stcurve, hazard- will give me what I want?
Any help will be much appreciated.
Thanks!
I'm running a Cox model using -stcox-. My main aim is to examine the effect of an independent continuous variable on the hazard to commit crime.
In the regression, the coefficient of this variable is positive and significant. However, I would like to "illustrate" the results and show what is the "predicted hazard" to commit crime, based on values of the independent variable. After linear regression, for example, I would use -margins- and get the predicted values of the dependent variable. But -margins- after -stcox- seems to give me something different.
So, my question is whether there is a good way to get predicted hazard to commit crime, i.e., to know what is the "risk" to commit crime during the follow-up period, based on values of my independent variable. Maybe using -stcurve, hazard- will give me what I want?
Any help will be much appreciated.
Thanks!
