| Title: | Two Stage Residual Inclusion Additive Hazards Estimator |
| Version: | 1.0.0 |
| Description: | Additive hazards models with two stage residual inclusion method are fitted under either survival data or competing risks data. The estimator incorporates an instrumental variable and therefore can recover causal estimand in the presence of unmeasured confounding under some assumptions. A.Ying, R. Xu and J. Murphy. (2019) <doi:10.1002/sim.8071>. |
| Depends: | R (≥ 3.5.0) |
| Imports: | survival |
| License: | LGPL-2 | LGPL-2.1 | LGPL-3 [expanded from: LGPL (≥ 2)] |
| Encoding: | UTF-8 |
| URL: | https://onlinelibrary.wiley.com/doi/abs/10.1002/sim.8071 |
| LazyData: | true |
| RoxygenNote: | 6.1.1 |
| NeedsCompilation: | no |
| Packaged: | 2020-04-23 20:06:44 UTC; andrewying |
| Author: | Andrew Ying [aut, cre] |
| Maintainer: | Andrew Ying <aying9339@gmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2020-04-28 10:50:02 UTC |
Plotting Predicted Survival Function or Cumulative Incidence Function with Pointwise Confidence Intervals
Description
The function will plot the predicted survival function when fitting a survival model and the predicted cumulative incidence function when fitting a competing risks model. Corresponding pointwise confidence intervals at level alpha are also included.
Usage
## S3 method for class 'tsriadditive'
plot(x, newtreatment = NULL, newIV = NULL,
newcovariates = NULL, alpha = 0.05, unit = "", ...)
Arguments
x |
the fitting object after fitting our model |
newtreatment |
a new treatment value |
newIV |
a new instrumental variable value |
newcovariates |
a new observed covariates |
alpha |
the confidence level 1 - alpha for confidence interval |
unit |
the time unit we focus |
... |
the other arguments you want to put in the built-in plot function |
Value
No return value, called for side effects
Examples
survtime <- rexp(100)
cause <- rbinom(100, 1, 0.7)
treatment <- rbinom(100, 1, 0.5)
IV <- rnorm(100)
covariates <- rnorm(100)
fit <- tsriadditive(survtime, cause, treatment, IV, covariates)
plot(fit, 1, 0, 0)
Predict method for Additive Hazards Model with Two Stage Residual Inclusion Method Fits
Description
Predicted values based on tsriadditive object.
Usage
## S3 method for class 'tsriadditive'
predict(object, newtreatment = NULL,
newIV = NULL, newcovariates = NULL, ...)
Arguments
object |
an object of class "tsriadditive", usually, a result of a call to tsriadditive. |
newtreatment |
a new treatment value. |
newIV |
a new instrumental variable value. |
newcovariates |
a new observed covariates. |
... |
further arguments passed to or from other methods. |
Value
predict.tsriadditive produces a venctor of predictions based on new values. A list with the following components is returned:
newobsz |
the vector grouping newtreatment, new IV and newcovariates |
score_pred |
the predicted scores |
hazard_pred |
the predicted baseline hazards function |
surival_pred |
the predicted surival function |
Examples
survtime <- rexp(100)
cause <- rbinom(100, 1, 0.7)
treatment <- rbinom(100, 1, 0.5)
IV <- rnorm(100)
covariates <- rnorm(100)
fit <- tsriadditive(survtime, cause, treatment, IV, covariates)
predict(fit, 1, 0, 0)
Summarizing Additive Hazards Model with Two Stage Residual Inclusion Method Fits
Description
summary method for class "tsriadditive".
Usage
## S3 method for class 'tsriadditive'
summary(object, ...)
## S3 method for class 'summary.tsriadditive'
print(x, ...)
Arguments
object |
an object of class "tsriadditive", usually, a result of a call to tsriadditive. |
... |
further arguments passed to or from other methods. |
x |
an object of class "summary.tsriadditive", usually, a result of a call to summary.tsriadditive. |
Value
print.summary.lm tries to be smart about formatting coefficients, an estimated variance covariance matrix of the coeffieients, Z-values and the corresponding P-values
Examples
survtime <- rexp(100)
cause <- rbinom(100, 1, 0.7)
treatment <- rbinom(100, 1, 0.5)
IV <- rnorm(100)
covariates <- rnorm(100)
fit <- tsriadditive(survtime, cause, treatment, IV, covariates)
summary(fit)
Fitting Additive Hazards Models with Two Stage Residual Inclusion Method
Description
tsriadditive is used to fit additive hazards models with two stage residual inclusion method.
Usage
tsriadditive(survtime, cause = NULL, treatment = NULL, IV = NULL,
covariates = NULL)
Arguments
survtime |
the event time |
cause |
the indicator records the cause. Default to all one. Zero means right censoring. Greater than or equal to two means other cause. |
treatment |
the treatment variable, can be null |
IV |
the instrumental variable |
covariates |
all the observed confounders |
Value
tsriadditive returns an object of class "tsriadditive". An object of class "tsriadditive" is a list containing the following components:
coef |
an estimate of the coefficients |
baseline |
an estimate of the baseline hazards function |
vcov |
an estimate of the variance covariance matrix of coef |
byprod |
a byproduct, that will used by other functions |
References
Ying, A., Xu, R. and Murphy, J. Two-Stage Residual Inclusion for Survival Data and Competing Risks - An Instrumental Variable Approach with Application to SEER- Medicare Linked Data. Statistics in Medicine, 38(10): 1775-1801, 2019.
Examples
survtime <- rexp(100)
cause <- rbinom(100, 1, 0.7)
treatment <- rbinom(100, 1, 0.5)
IV <- rnorm(100)
covariates <- rnorm(100)
fit <- tsriadditive(survtime, cause, treatment, IV, covariates)