fitit {icpack}

Fit proportional hazard model with smooth baseline hazard and (optional) interval censoring

Description

Fit proportional hazard model with smooth baseline hazard and (optional) interval censoring

Usage

fitit(
  Y,
  R,
  dead,
  X,
  B,
  Ic,
  R1,
  cbx,
  Pdiff,
  Pridge,
  lambda,
  nit = 50,
  tol = 1e-06,
  tollam = 0.01,
  update_lambda = FALSE,
  ic_update = TRUE,
  monitor = FALSE
)

Arguments

Y

Events (matrix, number of bins by subjects)

R

Risk sets (matrix, number of bins by subjects)

dead

(Boolean vector, TRUE if event, FALSE if right censored)

X

Covariates (matrix, number of covariates (+1) by subjects)

B

B-spline basis matrix

Ic

Censoring interval per individual, coded as 0/1 (in columns)

R1

Left truncation interval per individual, coded as 0/1 (in columns)

cbx

Vector of starting values

Pdiff

B-spline part of penalty matrix

Pridge

Ridge part of penalty matrix (for intercept)

lambda

Smoothing parameter (number)

nit

Maximum number of iterations (integer)

tol

Tolerance for final fit

tollam

Tolerance for switching to lambda update

update_lambda

Automatic update of lambda (Boolean)

ic_update

Update risk and event probabilities (Boolean)

monitor

Monitor convergence (Boolean)

Value

A list with items

cbx

Vector of

ll

Poisson GLM log-likelihood

lambda

Final tuning parameter

pen

Penalty part of penalized log-likelihood

ed

Effetive dimension of the baseline hazard

nit1

Number of iterations used in first phase

nit

Total number of iterations used (first plus second phase)

tollam

Tolerance used for switching to lambda update