Accelerated failure time regression

log T = x'beta + sigma W. Coefficients act MULTIPLICATIVELY ON TIME rather than on the hazard, so exp(beta) is a time ratio and not a hazard ratio.

Usage

Aftfit(time, event, X, dist = "weibull", alpha = 0.05)

morie_survival_aft(time, event, X, dist = "weibull", alpha = 0.05)

Arguments

time: observed follow-up times.
event: event indicator, 1 = event, 0 = censored.
X: covariate matrix, one row per subject.
dist: distribution of the error term.
alpha: significance level for the intervals.

Value

list with the coefficients, standard errors, time ratios, the scale and the log-likelihood.

References

Kalbfleisch, J. D. and Prentice, R. L. (2002). The Statistical Analysis of Failure Time Data, 2nd ed., Wiley, Chapter 3.

Examples

Aftfit(time = c(1, 2, 3, 4, 5, 6, 7, 8), event = c(0, 1, 0, 1, 1, 0, 1, 0),
  X = c(1, 2, 3, 4, 5, 6, 7, 8))
#> $dist
#> [1] "weibull"
#> 
#> $coef
#> [1] 0.4173392 0.2363297
#> 
#> $log_scale
#> [1] -2.733679
#> 
#> $scale
#> [1] 0.06497981
#> 
#> $loglik
#> [1] 2.494852
#> 
#> $n_par
#> [1] 3
#> 
#> $n
#> [1] 8
#> 
#> $n_events
#> [1] 4
#> 
#> $aic
#> [1] 1.010295
#> 
#> $bic
#> [1] 1.24862
#> 
#> $fixed_scale
#> [1] FALSE
#> 
#> $convergence
#> [1] 0
#> 
#> $intercept
#> [1] 0.4173392
#> 
#> $beta
#> [1] 0.2363297
#> 
#> $time_ratio
#> [1] 1.266592
#> 
#> $alpha
#> [1] 0.05
#> 
#> $positive_coef_means_longer_survival
#> [1] TRUE
#> 
#> $ph_coef
#> [1] -3.636971
#> 
#> $hazard_ratio
#> [1] 0.02633198
#> 
#> $ph_equivalent
#> [1] TRUE
#> 
#> $method
#> [1] "accelerated failure time, log T = x'beta + sigma W"
#>