Hi,
I would like to analyze multiple failure data with time-varying independent variable. I know how to analyze and set up the data for single failure event data with time-varying independent variable, and I know how to analyze and set up multiple failure time data with time-invariant independent variable. But, I have troubles in finding
a) whether it is possible to analyze multiple failure data with time-varying independent variable and
b) how I should set-up the dataset
More precisely, I am using Wei, Lin & Weissfeld method, and possible also Prentice, Williams and Peterson. My data includes 150 countries that are followed from 2002 until they complete so-called EITI process or until 2015 when the remaining countries are censored. The EITI process includes several phases and completion of each of these phases marks one of the multiple failures (so I have ordered failure events). My independent variables change over the period the countries are included in the dataset (for example, regime type and income per capita) and I would like to include that information in the analysis. So far, I have analyzed the data using stcox (WLW method), but keeping the independent variables time-invariant, set to their year 2001 value.
I would greatly appreciate if you could point me literature that helps me to understand how I should set up the database so that I could include the time varying information in the multiple failure survival analysis.
Thank you for your help in advance!
Best, Päivi
I would like to analyze multiple failure data with time-varying independent variable. I know how to analyze and set up the data for single failure event data with time-varying independent variable, and I know how to analyze and set up multiple failure time data with time-invariant independent variable. But, I have troubles in finding
a) whether it is possible to analyze multiple failure data with time-varying independent variable and
b) how I should set-up the dataset
More precisely, I am using Wei, Lin & Weissfeld method, and possible also Prentice, Williams and Peterson. My data includes 150 countries that are followed from 2002 until they complete so-called EITI process or until 2015 when the remaining countries are censored. The EITI process includes several phases and completion of each of these phases marks one of the multiple failures (so I have ordered failure events). My independent variables change over the period the countries are included in the dataset (for example, regime type and income per capita) and I would like to include that information in the analysis. So far, I have analyzed the data using stcox (WLW method), but keeping the independent variables time-invariant, set to their year 2001 value.
I would greatly appreciate if you could point me literature that helps me to understand how I should set up the database so that I could include the time varying information in the multiple failure survival analysis.
Thank you for your help in advance!
Best, Päivi