* using log directory ‘/data/localhost/ripley/R/packages/tests-MKL/CIMEHR.Rcheck’
* using R Under development (unstable) (2026-10-28 r90786)
* using platform: x86_64-pc-linux-gnu
* R was compiled by
    gcc (GCC) 16.2.1 20260819 (Red Hat 16.2.1-2)
    GNU Fortran (GCC) 16.2.1 20260819 (Red Hat 16.2.1-2)
* running under: Fedora Linux 44 (Server Edition)
* using session charset: UTF-8
* current time: 2026-10-29 09:58:28 UTC
* using option ‘--no-stop-on-test-error’
* checking for file ‘CIMEHR/DESCRIPTION’ ... OK
* checking extension type ... Package
* this is package ‘CIMEHR’ version ‘0.2.0’
* package encoding: UTF-8
* checking package namespace information ... OK
* checking package dependencies ... OK
* checking if this is a source package ... OK
* checking if there is a namespace ... OK
* checking for executable files ... OK
* checking for hidden files and directories ... OK
* checking for portable file names ... OK
* checking for sufficient/correct file permissions ... OK
* checking whether package ‘CIMEHR’ can be installed ... [76s/68s] OK
* used C++ compiler: ‘g++ (GCC) 16.2.1 20260819 (Red Hat 16.2.1-2)’
* checking C++ specification ... INFO
  specified C++17
* checking package directory ... OK
* checking ‘build’ directory ... OK
* checking DESCRIPTION meta-information ... OK
* checking top-level files ... OK
* checking for left-over files ... OK
* checking index information ... OK
* checking package subdirectories ... OK
* checking code files for non-ASCII characters ... OK
* checking R files for syntax errors ... OK
* checking whether the package can be loaded ... OK
* checking whether the package can be loaded with stated dependencies ... OK
* checking whether the package can be unloaded cleanly ... OK
* checking whether the namespace can be loaded with stated dependencies ... OK
* checking whether the namespace can be unloaded cleanly ... OK
* checking loading without being on the library search path ... OK
* checking use of S3 registration ... OK
* checking dependencies in R code ... OK
* checking S3 generic/method consistency ... OK
* checking replacement functions ... OK
* checking foreign function calls ... OK
* checking R code for possible problems ... [45s/52s] OK
* checking Rd files ... OK
* checking Rd metadata ... OK
* checking Rd line widths ... OK
* checking Rd cross-references ... OK
* checking for missing documentation entries ... OK
* checking for code/documentation mismatches ... OK
* checking Rd \usage sections ... OK
* checking Rd contents ... OK
* checking for unstated dependencies in examples ... OK
* checking contents of ‘data’ directory ... OK
* checking data for non-ASCII characters ... OK
* checking LazyData ... OK
* checking data for ASCII and uncompressed saves ... OK
* checking line endings in C/C++/Fortran sources/headers ... OK
* checking line endings in Makefiles ... OK
* checking compilation flags in Makevars ... OK
* checking for GNU extensions in Makefiles ... OK
* checking for portable use of $(BLAS_LIBS) and $(LAPACK_LIBS) ... OK
* checking use of PKG_*FLAGS in Makefiles ... OK
* checking use of SHLIB_OPENMP_*FLAGS in Makefiles ... OK
* checking pragmas in C/C++ headers and code ... OK
* checking compilation flags used ... OK
* checking compiled code ... OK
* checking installed files from ‘inst/doc’ ... OK
* checking files in ‘vignettes’ ... OK
* checking examples ... [9s/10s] OK
* checking for unstated dependencies in ‘tests’ ... OK
* checking tests ...
  Running ‘boundary-reporting.R’
  Running ‘ehrjoint-observation-covariates.R’
  Running ‘identification-diagnostics.R’
  Running ‘identification-warnings.R’ [18s/19s]
  Running ‘method-comparisons-arguments.R’
  Running ‘numerical-stability.R’
  Running ‘numerical-warnings.R’
  Running ‘pairwise-large-n.R’
  Running ‘pairwise-polishing.R’
  Running ‘zero-frailty-rank.R’
  Running ‘zero-frailty.R’
 [43s/48s] ERROR
Running the tests in ‘tests/ehrjoint-observation-covariates.R’ failed.
Complete output:
  > library(CIMEHR)
  > 
  > ehrjoint_fixture <- function() {
  +   set.seed(673)
  +   n <- 220L
  +   z <- rbinom(n, 1, .5); x <- rnorm(n); w <- rnorm(n); v <- rnorm(n)
  +   C <- runif(n, 2, 5); u <- rgamma(n, 1, 1)
  +   m <- rpois(n, 3*C*exp(.2*z-.15*x)*u); m[1L] <- 0L
  +   do.call(rbind, lapply(seq_len(n), function(i) {
  +     t <- if (m[i]) sort(runif(m[i], .01, C[i])) else NA_real_
  +     r <- if (m[i]) rbinom(m[i], 1, plogis(.3+.2*z[i]+.4*w[i]-.3*v[i])) else NA_integer_
  +     y <- 1+.1*t+.4*z[i]-.25*x[i]+.4*(u[i]-1)+rnorm(length(t))
  +     y[is.na(r) | r != 1] <- NA_real_
  +     data.frame(id=i, time=t, Y=y, R=r, C=C[i], Z=z[i], X=x[i], W=w[i], V=v[i])
  +   }))
  + }
  > 
  > local({
  +   d <- ehrjoint_fixture()
  +   args <- list(method='ehrjoint', data=d, visit_covs=c('Z','X'),
  +     long_covs=c('Z','X'), outcome_random_covs='Z', censor_col='C',
  +     controls=list(retain_influence=TRUE))
  +   default <- do.call(fit_method, args)
  +   explicit <- do.call(fit_method, c(args, list(obs_fixed_covs=c('Z','X'))))
  +   for (field in c('coefficients','se','vcov'))
  +     stopifnot(identical(default[[field]], explicit[[field]]))
  + 
  +   for (link in c('logit','probit')) {
  +     selected <- args
  +     selected$obs_fixed_covs <- c('Z','W','V')
  +     selected$controls$obs_link <- link
  +     fit <- do.call(fit_method, selected)
  +     visited <- d[is.finite(d$time),]
  +     reference <- glm(R ~ Z + W + V, visited, family=binomial(link=link))
  +     stopifnot(identical(names(fit$nuisance$alpha),names(coef(reference))),
  +       isTRUE(all.equal(unname(fit$nuisance$alpha),unname(coef(reference)),tolerance=1e-12)),
  +       identical(names(fit$nuisance$gamma),c('Z','X')),
  +       all(c('alpha:W','alpha:V') %in% colnames(fit$stacked_sandwich$bread)),
  +       !any(c('gamma:W','gamma:V') %in% colnames(fit$stacked_sandwich$bread)),
  +       identical(fit$refit$arguments$obs_fixed_covs,c('Z','W','V')),
  +       isTRUE(all.equal(crossprod(fit$stacked_sandwich$influence),
  +         fit$stacked_sandwich$vcov_joint,tolerance=1e-12)))
  +     idx <- paste0('outcome:',names(fit$coefficients))
  +     stopifnot(isTRUE(all.equal(unname(fit$vcov),
  +       unname(fit$stacked_sandwich$vcov_joint[idx,idx,drop=FALSE]),tolerance=1e-12)))
  +   }
  +   intercept <- do.call(fit_method,c(args,list(obs_fixed_covs=character())))
  +   stopifnot(identical(names(intercept$nuisance$alpha),'(Intercept)'),
  +     nrow(intercept$nuisance$vcov_alpha)==1L)
  +   changing <- d
  +   changing$W <- ifelse(is.finite(changing$time),changing$W+changing$time,changing$W)
  +   args$data <- changing
  +   bad <- try(do.call(fit_method,c(args,list(obs_fixed_covs=c('Z','W')))),silent=TRUE)
  +   stopifnot(inherits(bad,'try-error'),grepl('time-invariant',as.character(bad),fixed=TRUE))
  + })
  > 
  > local({
  +   # Recording-covariate defaults and overrides.
  +   d <- ehrjoint_fixture()
  +   args <- list(method='ehrjoint', data=d, visit_covs=c('Z','X'),
  +     long_covs=c('Z','X'), outcome_random_covs='Z', censor_col='C')
  +   same <- function(a, b) for (field in c('coefficients','se','vcov','nuisance'))
  +     stopifnot(identical(a[[field]], b[[field]]))
  +   overridden <- do.call(fit_method, c(args, list(controls=list(visit_covs='Z'))))
  +   reference <- do.call(fit_method, modifyList(args, list(visit_covs='Z')))
  +   same(overridden, reference)
  +   stopifnot(identical(names(overridden$nuisance$alpha), c('(Intercept)','Z')),
  +     identical(names(overridden$nuisance$gamma), 'Z'),
  +     identical(overridden$refit$arguments$obs_fixed_covs, 'Z'))
  +   same(do.call(EHRJoint, c(args[-1L], list(controls=list(visit_covs='Z')))), overridden)
  +   same(CIMEHR:::.curve_bootstrap_refit(overridden, seq_along(unique(d$id)), 1L), overridden)
  + 
  +   explicit <- do.call(fit_method, c(args, list(obs_fixed_covs=c('Z','X'),
  +     controls=list(visit_covs='Z'))))
  +   stopifnot(identical(names(explicit$nuisance$alpha), c('(Intercept)','Z','X')),
  +     identical(names(explicit$nuisance$gamma), 'Z'),
  +     !identical(explicit$coefficients, overridden$coefficients))
  +   same(do.call(fit_method, c(args, list(controls=list(visit_covs='Z',
  +     obs_fixed_covs=c('Z','X'))))), explicit)
  +   intercept <- do.call(fit_method, c(args, list(obs_fixed_covs=character(),
  +     controls=list(visit_covs='Z'))))
  +   stopifnot(identical(names(intercept$nuisance$alpha), '(Intercept)'))
  + })
  Error in same(overridden, reference) : 
    identical(a[[field]], b[[field]]) is not TRUE
  Calls: local ... eval.parent -> eval -> eval -> eval -> eval -> same -> stopifnot
  Execution halted
Running the tests in ‘tests/zero-frailty.R’ failed.
Complete output:
  > library(CIMEHR)
  > 
  > ns <- asNamespace("CIMEHR")
  > expect_identification_error <- function(expr, reason = NULL, stage = "outcome") {
  +   err <- tryCatch(force(expr), error = identity)
  +   stopifnot(inherits(err, "cimehr_zero_frailty_identification_error"),
  +     identical(err$status, "not_identified"), identical(err$stage, stage),
  +     identical(err$sigma2_zeta, 0), startsWith(conditionMessage(err),
  +       "This zero-frailty design cannot be uniquely estimated by the current three-stage procedure"))
  +   if (!is.null(reason)) stopifnot(identical(err$reason, reason))
  +   invisible(err)
  + }
  > 
  > for (variant in c("current", "higher_order_gaussian_moments")) {
  +   engine <- get(".cimehr_build_engine", ns)(variant)
  +   up <- engine$.cimehr_updated
  + 
  +   # Zero-information posterior variance.
  +   post <- up$.lu_laplace(c(0, 2, 5), c(0, 1.3, 4), 0)
  +   stopifnot(all(post$mu == 0), all(post$s2 == 1), all(post$mode == 0),
  +     all(post$curvature == 1), identical(post$sigma2_zeta, 0))
  +   if (is.function(engine$.cimehr_spline$.lu_laplace)) {
  +     spline_post <- engine$.cimehr_spline$.lu_laplace(c(0, 2, 5), c(0, 1.3, 4), 0)
  +     stopifnot(identical(post$mu, spline_post$mu), identical(post$s2, spline_post$s2))
  +   }
  + 
  +   # Scale-free compensation multiplier.
  +   Z <- c(-1, 0, 1)
  +   X <- c(-0.5, 0, 0.5)
  +   XO <- cbind("(Intercept)" = 1, X = X, Z = Z)
  +   ZO <- cbind("(Intercept)" = 1, Z = Z)
  +   par <- c("alpha.(Intercept)" = 0.2, "alpha.X" = 0.5, "alpha.Z" = 0.1,
  +     "delta.(Intercept)" = 0.7, "delta.Z" = -0.3,
  +     "sigma2_q.(Intercept)" = 0.4, "sigma2_q.Z" = 0.2)
  +   f <- up$.lu_functionals(par, XO, ZO, post, include_delta = TRUE)
  +   d <- 0.7 - 0.3 * Z
  +   total <- 1 + 0.4 + 0.2 * Z^2 + d^2
  +   k <- (0.2 + 0.5 * X + 0.1 * Z)/sqrt(total)
  +   h <- stats::dnorm(k)/stats::pnorm(k)/sqrt(total)
  +   stopifnot(max(abs(f$omega - stats::pnorm(k))) < 1e-14,
  +     max(abs(f$kappa - d*h)) < 1e-14, all(abs(f$kappa) > 0))
  +   attr(par, "zero_compensation") <- TRUE
  +   quotient <- up$.lu_functionals(par, XO, ZO, post, include_delta = TRUE)
  +   stopifnot(max(abs(quotient$kappa - h)) < 1e-14)
  + 
  +   n <- 48L
  +   z <- seq(-1.5, 1.5, length.out = n)
  +   x <- sin(seq_len(n))
  +   post_n <- up$.lu_laplace(rep(4, n), rep(4, n), 0)
  +   make_spec <- function(ZO, family, ...) {
  +     c(list(kind = "quotient", posterior = post_n, original_ZO = ZO,
  +       tolerance = 1e-7, family = family), list(...))
  +   }
  +   same_space <- function(A, B) {
  +     stopifnot(up$.lu_zero_rank(A) == up$.lu_zero_rank(B),
  +       up$.lu_zero_rank(cbind(A, B)) == up$.lu_zero_rank(A))
  +   }
  +   check_reuse <- function(result, W0, W1, ZO) {
  +     again <- up$.lu_zero_compensation(result$spec, W0, W1, ZO)
  +     stopifnot(identical(result$W0, again$W0), identical(result$W1, again$W1))
  +   }
  + 
  +   # One loading with an affine outcome random design.
  +   zi <- matrix(1, n, 1L)
  +   W0 <- cbind(1, x, 0)
  +   W1 <- cbind(rep(0,n), rep(0,n), rep(1,n))
  +   scalar <- make_spec(zi, "groups", profiles = matrix(1, 1L, 1L),
  +     group_variance = 0.8)
  +   ans <- up$.lu_zero_compensation(scalar, W0, W1)
  +   same_space(rbind(ans$W0, ans$W1), rbind(W0, W1))
  +   for (loading in c(-2, -0.3, 0.4, 3))
  +     same_space(rbind(ans$W0, ans$W1), loading*rbind(W0, W1))
  +   check_reuse(ans, W0, W1, zi)
  +   scalar$group_variance <- 0
  +   zero <- up$.lu_zero_compensation(scalar, W0, W1)
  +   stopifnot(ncol(zero$W0) == 0L, ncol(zero$W1) == 0L)
  + 
  +   # Saturated outcome group coefficients.
  +   g <- rep(0:1, length.out = n)
  +   zg <- cbind(1, g)
  +   Wg <- zg
  +   Wg1 <- matrix(0, n, ncol(Wg))
  +   groups <- make_spec(zg, "groups", profiles = rbind(c(1,0), c(1,1)),
  +     group_variance = c(0.6, 1.2))
  +   ans <- up$.lu_zero_compensation(groups, Wg, Wg1)
  +   same_space(ans$W0, zg)
  +   for (loading in list(c(-0.4,0.8), c(0.3,0.2), c(0.6,-1)))
  +     same_space(ans$W0, Wg*loading[g+1L])
  +   check_reuse(ans, Wg, Wg1, zg)
  + 
  +   # Common outcome random intercept with two recording loadings.
  +   expect_identification_error(up$.lu_zero_compensation(groups,
  +     matrix(1,n,1L), matrix(0,n,1L)),
  +     "compensation_direction_depends_on_unidentified_decomposition")
  + 
  +   # One zero-variance group and compensation c*(1-Z).
  +   one_group <- groups
  +   one_group$group_variance <- c(0.6, 0)
  +   a <- up$.lu_zero_compensation(one_group, matrix(1,n,1L), matrix(0,n,1L))
  +   np <- up$.lu_zero_compensation(a$spec, matrix(1,1L,1L), matrix(0,1L,1L),
  +     matrix(c(1,0.25),1L,2L))
  +   stopifnot(abs(np$W0[1L,1L]-0.75)<1e-12, np$W1[1L,1L]==0)
  + 
  +   # Continuous Z with a two-loading covariance ridge.
  +   zc <- cbind(1, z)
  +   edge <- make_spec(zc, "edge", axes = 1:2,
  +     reference_direction = c(0.6,-0.4))
  +   expect_identification_error(up$.lu_zero_compensation(edge, zc,
  +     matrix(0,n,2L)),
  +     "compensation_direction_depends_on_unidentified_decomposition")
  + 
  +   # Diagonal covariance with one or multiple positive axes.
  +   axes <- make_spec(zc, "axes", axes = 1L)
  +   ans <- up$.lu_zero_compensation(axes, matrix(1,n,1L), matrix(0,n,1L))
  +   same_space(ans$W0, matrix(1,n,1L))
  +   # Prediction at new continuous Z_O rows.
  +   new_w <- matrix(1, 1L, 1L)
  +   pred <- up$.lu_zero_compensation(ans$spec, new_w, 0*new_w,
  +     matrix(c(1, 0.125), 1L, 2L))
  +   stopifnot(max(abs(unname(pred$W0)-unname(new_w)))<1e-12)
  +   axes$axes <- 1:2
  +   expect_identification_error(up$.lu_zero_compensation(axes,
  +     matrix(1,n,1L), matrix(0,n,1L)),
  +     "compensation_direction_depends_on_unidentified_decomposition")
  +   # Nested compensation spaces for x3 in {-1, 0, 1}.
  +   x3 <- rep(c(-1, 0, 1), length.out = n)
  +   z3 <- cbind(1, x3, deparse.level = 0)
  +   W3 <- cbind(1, x3, x3^2, deparse.level = 0)
  +   nested <- up$.lu_zero_compensation(make_spec(z3, "axes", axes = 1:2), W3, 0*W3)
  +   same_space(nested$W0, W3)
  + 
  +   # Locally unique loading direction.
  +   raw <- make_spec(zc, "unique_up_to_common_sign")
  +   raw$kind <- "raw_unique"
  +   r <- up$.lu_zero_compensation(raw, zc, 0*zc)
  +   stopifnot(identical(r$W0, zc), identical(r$W1, 0*zc), identical(r$prefix, "theta."))
  + 
  +   # Prediction within and beyond the fitted design span.
  +   edge3 <- make_spec(z3, "edge", axes = 1:2, reference_direction = c(0.7, 0.3))
  +   ans <- up$.lu_zero_compensation(edge3, W3, 0*W3)
  +   same_space(ans$W0, W3)
  +   check_reuse(ans, W3, 0*W3, z3)
  +   again <- up$.lu_zero_compensation(ans$spec, W3[3:1, ], 0*W3[3:1, ], z3[3:1, ])
  +   stopifnot(identical(again$W0, ans$W0[3:1, , drop = FALSE]))
  +   for (x in c(0.5, 2))
  +     expect_identification_error(up$.lu_zero_compensation(ans$spec,
  +       matrix(c(1, x, x^2), 1L), matrix(0, 1L, 3L), matrix(c(1, x), 1L)),
  +       NULL)
  +   expect_identification_error(up$.lu_zero_compensation(ans$spec,
  +     matrix(c(1, 1, 0), 1L), matrix(0, 1L, 3L), matrix(c(1, 1), 1L)),
  +     NULL)
  + 
  +   # Three saturated patterns with exact indicator columns.
  +   level <- rep(1:3, length.out = n)
  +   zf <- cbind("(Intercept)" = 1, FA = +(level == 1), FB = +(level == 2))
  +   prof <- unique(zf); rownames(prof) <- NULL
  +   lab <- up$.lu_zero_pattern_labels(prof)
  +   Wf <- zf; colnames(Wf)[1L] <- "Intercept"
  +   three <- make_spec(zf, "groups", profiles = prof, labels = lab,
  +     group = match(do.call(paste, as.data.frame(zf)), do.call(paste, as.data.frame(prof))),
  +     group_variance = c(0.5, 0.7, 0.9))
  +   ans <- up$.lu_zero_compensation(three, Wf, 0*Wf)
  +   stopifnot(identical(lab, c("FA=1,FB=0", "FA=0,FB=1", "FA=0,FB=0")),
  +     identical(colnames(ans$W0), paste0("Intercept|", lab)),
  +     all(ans$W0 %in% c(0, 1)), all(colSums(ans$W0) > 0))
  +   for (loading in list(c(1, 1, 1), c(-0.4, 0.8, 2), c(0.3, -0.2, 1)))
  +     same_space(ans$W0, Wf*loading[level])
  +   check_reuse(ans, Wf, 0*Wf, zf)
  +   expect_identification_error(up$.lu_zero_compensation(ans$spec,
  +     Wf[1:2, ], 0*Wf[1:2, ], rbind(zf[1, ], c(1, 1, 1))),
  +     NULL)
  + 
  +   # Baseline profiling with constant compensation.
  +   id <- rep(seq_len(n), each = 4L)
  +   dat <- data.frame(id=id, time=rep(c(0.1,0.3,0.6,0.9),n),
  +     R=1, C=1, Z=rep(z,each=4L))
  +   dat$Y <- 0.2 + 0.3*dat$Z + dat$time + sin(id)/10
  +   check <- up$.lu_validate_data(dat, "id", "time", "Y", "R", "C", "Z")
  +   XV <- up$.lu_design(dat, check$sf, "Z")
  +   stage1 <- up$.cimehr_shared_engine$.cimehr_fit_visiting(dat$time,
  +     check$visit, check$subject, dat$C[check$sf], XV,
  +     time_precision=0, score_tol=1e-10, Ei_subj=check$Ei)
  +   scalar$group_variance <- 0.8
  +   expect_identification_error(up$.lu_stage3(dat, check$sf, check$subject,
  +     dat$time, dat$Y, dat$R, check$visit, stage1, "Z", character(), "time",
  +     omega=rep(0.5,n), kappa=rep(0.4,n), include_theta=TRUE,
  +     zero_spec=scalar), "compensation_not_separable_from_outcome")
  + }
  > 
  > # Fitted zero-frailty designs.
  > sim_zero <- function(n, seed, sd_logU = 0, w_effect = 0) {
  +   set.seed(seed + 1000); W <- rnorm(n)
  +   set.seed(seed)
  +   Z <- rep(c(0, 1), length.out = n); X <- rnorm(n)
  +   logU <- if (sd_logU > 0) rnorm(n, -.5 * sd_logU^2, sd_logU) else rep(0, n)
  +   q <- rnorm(n, 0, .9)
  +   b <- .35 * logU + .4 * q + rnorm(n, 0, .6)
  +   m <- rpois(n, 10 * exp(.25 * Z - .15 * X) * exp(logU))
  +   id <- rep(seq_len(n), pmax(m, 1L)); has <- m[id] > 0
  +   N <- length(id); tt <- rep(NA_real_, N)
  +   tt[has] <- runif(sum(has), .025, 4.975)
  +   o <- order(id, tt); id <- id[o]; tt <- tt[o]; has <- has[o]
  +   rr <- rep(NA_integer_, N); yy <- rep(NA_real_, N)
  +   k <- sum(has); ii <- id[has]
  +   rr[has] <- as.integer(rnorm(k) < .25 + .2 * Z[ii] - .1 * X[ii] + .55 * logU[ii] + q[ii] + w_effect * W[ii])
  +   yy[has] <- 1 + .8 * Z[ii] - .4 * X[ii] + .2 * tt[has] + b[ii] + rnorm(k, 0, .7)
  +   yy[has & rr == 0L] <- NA_real_
  +   d <- data.frame(id = id, time = tt, Y = yy, R = rr, Z = Z[id], X = X[id], C = 5)
  +   d$W <- W[d$id]
  +   d
  + }
  > # Numerically identical covariates with different names.
  > dz <- sim_zero(300, 12)
  > dz$Xalias <- dz$X
  > set.seed(20261013); level <- sample(1:3, 300L, TRUE)
  > dz$FA <- +(level[dz$id] == 1); dz$FB <- +(level[dz$id] == 2); dz$FC <- +(level[dz$id] == 3)
  > # Recording influenced by the exclusion variable W.
  > dw <- sim_zero(300, 12, w_effect = 1)
  > dw[c('FA', 'FB', 'FC')] <- dz[c('FA', 'FB', 'FC')]
  > methods <- c('cimehr', 'cimehr_hl', 'cimehr_spline', 'cimehr_spline_hl')
  > zero_fit <- function(method, data = dz, ...) do.call(fit_method, c(list(method = method,
  +   data = data), utils::modifyList(list(visit_covs = c('Z', 'X'), long_covs = c('Z', 'X'),
  +   censor_col = 'C', seed = 1, controls = list(pairwise_control = list(score_tol = 1e-7))),
  +   list(...))))
  > frailty_variance <- function(f) if (grepl('spline', f$method_key))
  +   f$fit$nuisance_summary$visiting_process$sigma2_zeta else unname(coef(f)['sigma2_zeta'])
  > intercept_name <- function(method) if (grepl('spline', method)) '(Intercept)' else 'Intercept'
  > 
  > for (method in methods) {
  +   # Equal outcome and recording fixed designs.
  +   f <- suppressWarnings(zero_fit(method))
  +   alias <- suppressWarnings(zero_fit(method, obs_fixed_covs = c('Z', 'Xalias')))
  +   stopifnot(identical(frailty_variance(f), 0))
  +   comp <- paste0('compensation.', intercept_name(method))
  +   stopifnot(!comp %in% names(coef(f)), !any(grepl('group|[.]c[0-9]', names(coef(f)))),
  +     identical(f$diagnostics$zero_frailty_compensation$status, 'omitted_no_relevant_exclusion'),
  +     identical(alias$diagnostics$zero_frailty_compensation, f$diagnostics$zero_frailty_compensation))
  +   out <- summary_outcome(f)
  +   out_alias <- summary_outcome(alias)
  +   stopifnot(all(out$point_valid), all(out$se_available),
  +     all(is.finite(out$lower)), all(is.finite(out$upper)),
  +     identical(out$estimate, out_alias$estimate), identical(out$se, out_alias$se),
  +     identical(out$lower, out_alias$lower), identical(out$upper, out_alias$upper),
  +     identical(out$ci_available, out_alias$ci_available),
  +     all(is.finite(confint(f))))
  +   if (grepl('spline', method)) stopifnot(any(startsWith(out$parameter, 'baseline.')))
  +   obs <- summary_observation(f)
  +   stopifnot(identical(obs$parameter, c('alpha.(Intercept)', 'alpha.Z', 'alpha.X', 'tau2.(Intercept)')),
  +     all(is.finite(population_trajectory(f, times = c(1, 2))$fit)),
  +     all(is.finite(baseline_mean(f, times = c(1, 2))$fit)))
  +   if (grepl('spline', method)) {
  +     stopifnot(all(is.finite(population_trajectory(f, times = c(1, 2), se.fit = TRUE)$se)),
  +       all(is.finite(baseline_mean(f, times = c(1, 2), se.fit = TRUE)$se)))
  +   }
  + 
  +   # Binary Z across exclusion-restriction settings.
  +   f <- zero_fit(method, data = dw, obs_fixed_covs = c('Z', 'X', 'W'),
  +     obs_random_covs = 'Z', outcome_random_covs = 'Z')
  +   int <- intercept_name(method)
  +   stopifnot(identical(f$diagnostics$zero_frailty_compensation$status, 'estimated'),
  +     identical(summary_observation(f)$parameter[5:6], c('tau2.Z=0', 'tau2.Z=1')),
  +     identical(grep('^compensation[.]', names(coef(f)), value = TRUE),
  +       paste0('compensation.', int, '|Z=', 0:1)),
  +     all(is.finite(summary_outcome(f)$lower)), all(is.finite(summary_outcome(f)$upper)))
  +   f <- suppressWarnings(zero_fit(method, obs_random_covs = 'Z', outcome_random_covs = 'Z'))
  +   stopifnot(identical(f$diagnostics$zero_frailty_compensation$status, 'omitted_no_relevant_exclusion'),
  +     identical(summary_observation(f)$parameter[4:5], c('tau2.Z=0', 'tau2.Z=1')),
  +     !any(grepl('^compensation[.]', names(coef(f)))),
  +     all(is.finite(summary_outcome(f)$lower)), all(is.finite(summary_outcome(f)$upper)))
  + 
  +   # Factor levels with redundant dummy columns.
  +   f <- zero_fit(method, data = dw, obs_fixed_covs = c('Z', 'X', 'W'),
  +     obs_random_covs = c('FA', 'FB'), outcome_random_covs = c('FA', 'FB'))
  +   g <- zero_fit(method, data = dw, obs_fixed_covs = c('Z', 'X', 'W'),
  +     obs_random_covs = c('FA', 'FB', 'FC'), outcome_random_covs = c('FA', 'FB'))
  +   comp <- grep('^compensation[.]', names(coef(f)), value = TRUE)
  +   stopifnot(identical(comp, paste0('compensation.', int, '|',
  +       c('FA=0,FB=0', 'FA=1,FB=0', 'FA=0,FB=1'))),
  +     identical(grep('^tau2[.]', summary_observation(f)$parameter, value = TRUE),
  +       c('tau2.FA=0,FB=0', 'tau2.FA=1,FB=0', 'tau2.FA=0,FB=1')),
  +     identical(unname(coef(f)[c('beta.Z', 'beta.X', comp)]),
  +       unname(coef(g)[grepl('^(beta|compensation)[.]', names(coef(g)))])),
  +     identical(summary_outcome(f)$se, summary_outcome(g)$se))
  + }
  Error: identical(out$estimate, out_alias$estimate) is not TRUE
  Execution halted
* checking for unstated dependencies in vignettes ... OK
* checking package vignettes ... OK
* checking re-building of vignette outputs ... [16s/18s] OK
* checking PDF version of manual ... OK
* checking HTML version of manual ... OK
* checking for non-standard things in the check directory ... OK
* checking for detritus in the temp directory ... OK
* DONE

Status: 1 ERROR
See
  ‘/data/localhost/ripley/R/packages/tests-MKL/CIMEHR.Rcheck/00check.log’
for details.

Command exited with non-zero status 1
Time 4:05.07, 211.60 + 12.66