`CoxMaster`

objects instantiate and run a distributed Cox model
computation fit

1 |

An `R6Class`

generator object

`CoxMaster$new(defnId, formula, debug=FALSE)`

Create a new CoxMaster object using the defnId and formula. The debug flag is useful for debugging

`logLik(beta, ...)`

Compute the partial log likelihood for all the data by aggregating the values at each site. The return value is numeric scalar with two attributes:

`gradient`

contains the score vector, and`hessian`

contains the estimated hessian matrix`addSite(name, url)`

Add a worker site for participating in the distributed computation

`var(beta, ...)`

Compute the variance of the parameter vector beta

`kosher()`

Check if inputs and state of object are sane. For future use

`getP()`

Returns the dimension of the parameter vector

`run()`

Run the fitting iterations and save the result

`summary()`

Return a summary data frame columns for

`coef`

,`exp(coef)`

, standard error, z-score, and p-value for each parameter in the model following the same format as the`survival`

package

`CoxWorker`

which generates objects matched to such a master object

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.