The SURVEYPHREG Procedure

The hazard ratio for a quantitative effect with regression coefficient is defined as exp, where denotes the jth unit vector. In general, a log-hazard ratio can be written as , a linear combination of the regression coefficients, and the hazard ratio exp is obtained by replacing with .

The confidence intervals for hazard ratios are obtained by exponentiating the confidence limits of the corresponding linear combination. Thus, the confidence limits are

where is the percentile point of the t distribution with df degrees of freedom. For more information about hazard ratios see the section Hazard Ratios, and for more information about df see the section Degrees of Freedom. If you use the DF=NONE option in the MODEL statement, then the procedure uses the percentile point of the standard normal distribution.