The input table can be partitioned by either name or id and the hazard ratio is determined. Because reference values are not used in this example, partitioning by id gives the same result as identity comparison (that is, the patients are compared with themselves and the hazard ratio is one).