Model predictions for grpsurv objects

Similar to other predict methods, this function returns predictions from a fitted grpsurv object.

Usage

# S3 method for grpsurv
predict(
  object,
  X,
  type = c("link", "response", "survival", "hazard", "median", "norm", "coefficients",
    "vars", "nvars", "groups", "ngroups"),
  lambda,
  which = 1:length(object$lambda),
  ...
)

Arguments

object

Fitted grpsurv model object.

X

Matrix of values at which predictions are to be made. Not required for some type values.

type

Type of prediction:

link: linear predictors
response: risk (i.e., exp(link))
survival: the estimated survival function
hazard: the estimated cumulative hazard function
median: median survival time
The other options are all identical to their grpreg() counterparts

lambda

Regularization parameter at which predictions are requested. For values of lambda not in the sequence of fitted models, linear interpolation is used.

which

Indices of the penalty parameter lambda at which predictions are required. Default: all indices. If lambda is specified, this will override which.

...

Not used.

Value

The object returned depends on type.

Details

Estimation of baseline survival function conditional on the estimated values of beta is carried out according to the method described in Chapter 4.3 of Kalbfleisch and Prentice.

References

Kalbfleish JD and Prentice RL (2002). The Statistical Analysis of Failure Time Data, 2nd edition. Wiley.

Author

Patrick Breheny

Examples

data(Lung)
X <- Lung$X

y <- Lung$y
group <- Lung$group
 
fit <- grpsurv(X, y, group)
coef(fit, lambda=0.05)
#>         trt      karno1      karno2      karno3   diagtime1   diagtime2 
#>  0.08037515 -6.17591904  0.78016422 -0.46524844  0.00000000  0.00000000 
#>        age1        age2        age3       prior    squamous       small 
#> -0.30698476  0.28311172 -0.98219083  0.00000000 -0.39419164  0.17559226 
#>       adeno       large 
#>  0.41116562 -0.18990837 
head(predict(fit, X, type="link", lambda=0.05))
#>          1          2          3          4          5          6 
#> -0.3797934 -0.5934333 -0.2596129 -0.3895304 -0.5881482  0.7865260 
head(predict(fit, X, type="response", lambda=0.05))
#>         1         2         3         4         5         6 
#> 0.6840027 0.5524274 0.7713501 0.6773749 0.5553547 2.1957550 
 
# Survival function
S <- predict(fit, X[1,], type="survival", lambda=0.05)
S(100)
#> [1] 0.5126102
S <- predict(fit, X, type="survival", lambda=0.05)
plot(S, xlim=c(0,200))

 
# Medians
predict(fit, X[1,], type="median", lambda=0.05)
#> [1] 105
M <- predict(fit, X, type="median")
M[1:10, 1:10]
#>       [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
#>  [1,]   80   80   80   73   80   80   84   87   90    92
#>  [2,]   80   80   80   82   84   90   95   95   99   100
#>  [3,]   80   80   80   73   80   80   84   87   90    92
#>  [4,]   80   80   80   73   80   80   84   87   90    92
#>  [5,]   80   80   80   82   84   90   95   95   99   100
#>  [6,]   80   61   54   52   51   48   45   43   42    36
#>  [7,]   80   72   61   59   56   56   56   56   56    56
#>  [8,]   80   82   87   90   95   99  103  110  111   112
#>  [9,]   80   73   72   63   63   72   72   73   80    80
#> [10,]   80   80   80   82   84   90   95   95   99   100
 
# Nonzero coefficients
predict(fit, type="vars", lambda=c(0.1, 0.01))
#> $`0.1`
#>   karno1   karno2   karno3 squamous    small    adeno    large 
#>        2        3        4       11       12       13       14 
#> 
#> $`0.01`
#>       trt    karno1    karno2    karno3 diagtime1 diagtime2      age1      age2 
#>         1         2         3         4         5         6         7         8 
#>      age3     prior  squamous     small     adeno     large 
#>         9        10        11        12        13        14 
#> 
predict(fit, type="nvars", lambda=c(0.1, 0.01))
#>  0.1 0.01 
#>    7   14