NEWS

drgee 1.1.10 (2020-01-09)

Fixed compatibility with R version 4, by exchanging uses of the function class with the function inherits in conjunction with try. Also repleced class by inherits when checking the class of objects.
Changed Depends to Imports. This means that the packages nleqslv, survival, Rcpp and data.table are no longer automatically loaded when drgee is loaded.

Fixed a bug that caused R to crash when the clustering variable was a factor.
The object returned from gee now contains the original formula object and the original dataset
The object returned from drgee now contains the original formula objects and the original dataset

Fixed a bug regarding the ordering of observations in the data returned from gee and drgee

Updated the contact information for Johan Zetterqvist.
The functions drgee and gee now returns U - the residuals from the estimating equations evaluated at the estimated values of the parameters. They also return the matrix d.U.sum - the sum of the derivatives of the estimating equations evaluated at the estimated values of the parameters.
A function robustVcov takes U, d.U.sum and an optional clustervariable as arguments and returns the asymptotic variance matrix.
It is now possible to use a subset argument in the functions drgee and gee
For conditional gee, it is now possible to adjust for another clustervariable than the one used in the analysis (assuming) that the old clustervariable is contained within the new

Added a helper function to obtain score residuals from clogit given the estimated coefficients, the observed outcomes, the design matrix and the id vector.
Fixed a bug when using retrospective conditional logistic regression
Add doubly robust estimation in conditional logistic models

In some data manipulation steps, the data.table package is used to obtain better speed.
For estimation in conditional logistic models only outcome-discordant clusters are used
Calculation of residuals for estimating equations for conditional logistic models implemented in C++ using the Rcpp and RcppArmadillo packages.
Drops unused levels for factor variables

Fixed a bug in 'drgeeData' which caused an error when one of the nuisance models only contained an intercept

Fixed a bug in 'drgeeData' for the 'estimation.method' which happened when 'cond = TRUE' and 'olink = "logit"'.

Fixed a bug in 'drgeeData' that caused error when no data argument is supplied.
There was an error in the previous version, when the function 'drgeeData' removed outcome concordant cluster for conditional methods. This is now corrected such that this only happens for conditional logistic methods.

The arguments 'outcome', 'exposure' and 'clusterid' can be supplied as vectors or as a string.
Changed the interface for the 'drgee' function. Now the argument 'estimationMethod' has to be "dr","o", or "e" corresponding to previous choices "dr", "obe" and "ebe". When 'estimationMethod="o"', the user can supply an argument 'exposure' instead of 'emodel'. Similarly, when 'estimationMethod="e"', the user can supply an argument 'outcome' instead of 'omodel'.
Changed the functions 'drgee' and 'gee' such that they always return the variance as a matrix.
Added a function 'gee' which calculates all coefficients in a regression.
Added an option 'cond' for 'drgee' and 'gee' for models with cluster-specific intercepts.

Fixed a bug in 'drFit'. The calculation of the Jacobian was incorrect for doubly robust estimation with outcome link logit, resulting in inconsistent estimates of standard errors. This bug is now fixed.