Cox regression

Called by coxreg, but a user can call it directly.

coxreg.fit(
  X,
  Y,
  rs,
  weights,
  t.offset = NULL,
  strats,
  offset,
  init,
  max.survs,
  method = "efron",
  boot = FALSE,
  efrac = 0,
  calc.martres = TRUE,
  control,
  verbose = TRUE,
  calc.hazards = NULL,
  center = NULL
)

Arguments

X	The design matrix.
Y	The survival object.
rs	The risk set composition. If absent, calculated.
weights	Case weights; time-fixed or time-varying.
t.offset	Case offset; time-varying.
strats	The stratum variable. Can be absent.
offset	Offset. Can be absent.
init	Start values. If absent, equal to zero.
max.survs	Sampling of risk sets? If so, gives the maximum number of survivors in each risk set.
method	Either "efron" (default) or "breslow".
boot	Number of bootstrap replicates. Defaults to FALSE, no bootstrapping.
efrac	Upper limit of fraction failures in 'mppl'.
calc.martres	Should martingale residuals be calculated?
control	See coxreg
verbose	Should Warnings about convergence be printed?
calc.hazards	Deprecated. See coxreg.
center	Deprecated. See coxreg.

Value

A list with components

coefficients

Estimated regression parameters.

var

Covariance matrix of estimated coefficients.

loglik

First component is value at init, second at maximum.

score

Score test statistic, at initial value.

linear.predictors

Linear predictors.

residuals

Martingale residuals.

hazard

Estimated baseline hazard. At value zero of 'design' variables.

means

Means of the columns of the design matrix.

bootstrap

The bootstrap replicates, if requested on input.

conver

TRUE if convergence.

f.conver

TRUE if variables converged.

fail

TRUE if failure.

iter

Number of performed iterations.

Details

rs is dangerous to use when NA's are present.

Note

It is the user's responsibility to check that indata is sane.

See also

coxreg, risksets

Author

Göran Broström

Examples

 X <- as.matrix(data.frame(
                x=     c(0, 2,1,4,1,0,3),
                sex=   c(1, 0,0,0,1,1,1)))
 time <- c(1,2,3,4,5,6,7)
 status <- c(1,1,1,0,1,1,0)
 stratum <- rep(1, length(time))

 coxreg.fit(X, Surv(time, status), strats = stratum, max.survs = 6,
     control = list(eps=1.e-4, maxiter = 10, trace = FALSE))
#> $coefficients
#> [1] -0.7716698 -1.7543183
#> 
#> $df
#> [1] 2
#> 
#> $n
#> [1] 7
#> 
#> $sd
#> [1] 0.5529464 1.3474828
#> 
#> $var
#>           [,1]      [,2]
#> [1,] 0.3057497 0.3037635
#> [2,] 0.3037635 1.8157100
#> 
#> $loglik
#> [1] -7.138867 -5.273533
#> 
#> $score
#> [1] 3.274765
#> 
#> $linear.predictors
#>            [,1]
#> [1,]  0.4607732
#> [2,]  0.6717520
#> [3,]  1.4434217
#> [4,] -0.8715875
#> [5,] -0.3108965
#> [6,]  0.4607732
#> [7,] -1.8542360
#> 
#> $residuals
#> [1]  0.8470621  0.5829125 -0.7145486 -0.1693379  0.4243571 -1.4838069 -0.2453139
#> 
#> $noOfRisksets
#> [1] 5
#> 
#> $hazards
#> [[1]]
#>      [,1]      [,2]
#> [1,]    1 0.8839056
#> [2,]    2 1.0681447
#> [3,]    3 1.7571622
#> [4,]    5 3.4881900
#> [5,]    6 7.1577751
#> 
#> attr(,"class")
#> [1] "hazdata"
#> 
#> $means
#>         x       sex 
#> 1.5714286 0.5714286 
#> 
#> $bootstrap
#> NULL
#> 
#> $boot.sd
#> NULL
#> 
#> $conver
#> [1] 1
#> 
#> $f.conver
#> [1] 1
#> 
#> $fail
#> [1] 0
#> 
#> $iter
#> [1] 5
#>