Merge pull request #292 from florianhartig/0.4.3-glmmTMB-issue

0.4.3 glmm tmb issue

Author: florianhartig

.travis.yml

@@ -17,10 +17,8 @@ sudo: false
 # cache: packages
 
 r:
-  - oldrel
   - release
-  - devel
-  
+
 # Install nlopt package to help nloptr install
 # glpk-utils and libglpk-dev are needed because of DHARMa #166
 addons:

DHARMa/tests/testthat/testModelTypes.R Code/testModelTypes.R

@@ -310,13 +310,13 @@ test_that("glm poisson works",
             # fittedModel2 <- glm.nb(observedResponse ~ Environment1, data = testData2)
             # expectDispersion(fittedModel2,FALSE)
 
-            fittedModel <- glmmTMB(observedResponse ~ Environment1 , family = "poisson", data = testData)
-            runEverything(fittedModel, testData)
-            fittedModel2 <- glmmTMB(observedResponse ~ Environment1 , family = "poisson", data = testData2)
-            expectDispersion(fittedModel2)
+            #fittedModel <- glmmTMB(observedResponse ~ Environment1 , family = "poisson", data = testData)
+            #runEverything(fittedModel, testData)
+            #fittedModel2 <- glmmTMB(observedResponse ~ Environment1 , family = "poisson", data = testData2)
+            #expectDispersion(fittedModel2)
 
-            fittedModel <- glmmTMB(observedResponse ~ Environment1 + (1|group), zi=~1 , family = nbinom2, data = testData)
-            fittedModel2 <- glmmTMB(observedResponse ~ Environment1 + (1|group), zi=~1 , family = nbinom2, data = testData2)
+            #fittedModel <- glmmTMB(observedResponse ~ Environment1 + (1|group), zi=~1 , family = nbinom2, data = testData)
+            #fittedModel2 <- glmmTMB(observedResponse ~ Environment1 + (1|group), zi=~1 , family = nbinom2, data = testData2)
             # does not fully work
             # runEverything(fittedModel, testData)
             # expectDispersion(fittedModel2, F)
@@ -389,8 +389,8 @@ test_that("glm poisson weights throws warning",
                       expect_warning(simulateResiduals(fittedModel))
 
                       # glmmTMB does not warn, implemented warning in DHARMa
-                      fittedModel <- glmmTMB(observedResponse ~ Environment1 , family = "poisson", data = testData, weights = weights)
-                      expect_warning(simulateResiduals(fittedModel))
+                      # fittedModel <- glmmTMB(observedResponse ~ Environment1 , family = "poisson", data = testData, weights = weights)
+                      # expect_warning(simulateResiduals(fittedModel))
 
                       # spaMM does not warn, but seems to be simulating with correct (heteroskedastic) variance.
                       # weights cannot be fit with poisson, because spaMM directly interpretes weights as variance

DHARMa/DESCRIPTION

@@ -1,7 +1,7 @@
 Package: DHARMa
 Title: Residual Diagnostics for Hierarchical (Multi-Level / Mixed) Regression Models
-Version: 0.4.2
-Date: 2021-04-8
+Version: 0.4.3
+Date: 2021-07-06
 Authors@R: c(person("Florian", "Hartig", email = "florian.hartig@biologie.uni-regensburg.de", role = c("aut", "cre"), comment=c(ORCID="0000-0002-6255-9059")), person("Lukas", "Lohse", role = "ctb"))
 Description: The 'DHARMa' package uses a simulation-based approach to create
     readily interpretable scaled (quantile) residuals for fitted (generalized) linear mixed
@@ -28,8 +28,7 @@ Imports:
     lmtest,
     ape,
     qgam (>= 1.3.2),
-    lme4,
-    glmmTMB (>= 1.0.1)
+    lme4
 Suggests:
     knitr,
     testthat,

DHARMa/NEWS

@@ -1,5 +1,13 @@
 NOTE: for more news about the package, see https://github.com/florianhartig/DHARMa/releases
 
+DHARMa 0.4.3
+============
+
+Bugfixes
+
+- Removed glmmTMB completely, see #289
+
+
 DHARMa 0.4.2
 ============
 

DHARMa/R/compatibility.R

@@ -168,7 +168,7 @@ getFixedEffects <- function(fittedModel){
   } else if(class(fittedModel)[1] %in% c("glmerMod", "lmerMod", "HLfit")){
     out = fixef(fittedModel)
   } else if(class(fittedModel)[1] %in% c("glmmTMB")){
-    out = glmmTMB::fixef(fittedModel)
+    out = fixef(fittedModel)
     out = out$cond
   } else {
     out = coef(fittedModel)
@@ -232,7 +232,7 @@ getFitted <- function (object, ...) {
 #' @export
 getFitted.default <- function (object,...){
   out = predict(object, type = "response", re.form = ~0)
-  out = as.vector(out)
+  out = as.vector(out) # introduced because of phyr error
 }
 
 

DHARMa/cran-comments.md

@@ -1,4 +1,19 @@
-# Version 0.4.1
+# Version 0.4.3
+
+## Submission 1, 6.7.2021
+
+This is a bugfix release, attempting to fix the issue with the dependency on glmmTMB
+
+See NEWS for details. 
+
+This release was tested without apparent problems under
+
+* local MAC OS 11.4 (Big Sur), R 4.1.0
+* http://win-builder.r-project.org/ - oldrelease / devel / release
+* Linux (Travis CI) - release
+* GitHub Actions - Windows / Mac / Ubuntu (release), Ubuntu (devel)
+
+# Version 0.4.2
 
 ## Submission 1, 5.7.2021
 

DHARMa/inst/examples/hurricanes.R

@@ -1,8 +1,9 @@
+\dontrun{
 # Loading hurricanes dataset
 
 library(DHARMa)
-data(hurricanes)
 
+data(hurricanes)
 str(hurricanes)
 
 # this is the model fit by Jung et al.
@@ -24,4 +25,5 @@ res2 = recalculateResiduals(res, group = hurricanes$Year)
 testTemporalAutocorrelation(res2, time = unique(hurricanes$Year))
 
 # task: try to address these issues - in many instances, this will 
-# make the MasFem predictor n.s.
+# make the MasFem predictor n.s.
+}

DHARMa/man/hurricanes.Rd

@@ -367,8 +367,8 @@ testZeroInflation(simulationOutput)
 
 This test is likely better suited for detecting zero-inflation than the standard plot, but note that also overdispersion will lead to excess zeros, so only seeing too many zeros is not a reliable diagnostics for moving towards a zero-inflated model. A reliable differentiation between overdispersion and zero-inflation will usually only be possible when directly comparing alternative models, e.g. through residual comparison / model selection of a model with / without zero-inflation, or by simply fitting a model with zero-inflation and looking at the parameter estimate for the zero-inflation. A good option is the R package glmmTMB, which is also supported by DHARMa. We can use this to fit
 
-```{r}
-library(glmmTMB)
+```{r, eval= F}
+# requires glmmTMB
 fittedModel <- glmmTMB(observedResponse ~ Environment1 + I(Environment1^2) + (1|group), ziformula = ~1 , family = "poisson", data = testData)
 summary(fittedModel)
 
@@ -565,28 +565,26 @@ I guess you could use either Model 2 or 3 - the broader point is: increasing mod
 
 ## Owl example (count data)
 
-The next examples uses the fairly well known Owl dataset which is provided in glmmTMB (see ?Owls for more info about the data).
+The next examples uses the fairly well known Owl dataset which is provided in glmmTMB (see ?Owls for more info about the data). The following shows a sequence of models, all checked with DHARMa. The example is discussed in a talk at ISEC 2018, see slides [here](https://www.slideshare.net/florianhartig/mon-c5hartig2493). 
 
-The following shows a sequence of models, all checked with DHARMa. The example is discussed in a talk at ISEC 2018, see slides [here](https://www.slideshare.net/florianhartig/mon-c5hartig2493). 
 
-
-```{r}
+```{r, eval = F}
 m1 <- glm(SiblingNegotiation ~ FoodTreatment*SexParent + offset(log(BroodSize)), data=Owls , family = poisson)
 res <- simulateResiduals(m1)
 plot(res)
 ```
 
 OK, this is highly overdispersed. Let's add a RE on nest
 
-```{r}
+```{r, eval = F}
 m2 <- glmer(SiblingNegotiation ~ FoodTreatment*SexParent + offset(log(BroodSize)) + (1|Nest), data=Owls , family = poisson)
 res <- simulateResiduals(m2)
 plot(res)
 ```
 
 Somewhat better, but not good. Move to neg binom, to adjust dispersion
 
-```{r}
+```{r, eval= F}
 m3 <- glmmTMB(SiblingNegotiation ~ FoodTreatment*SexParent + offset(log(BroodSize)) + (1|Nest), data=Owls , family = nbinom1)
 
 res <- simulateResiduals(m3, plot = T)
@@ -598,7 +596,7 @@ testZeroInflation(res)
 
 We see underdispersion now. In a model with variable dispersion, this is often the signal that some other distributional assumptions are violated, that is why I checked for zero-inflation, and it looks as if there is some. Therefore fitting a zero-inflated model
 
-```{r}
+```{r, eval= F}
 m4 <- glmmTMB(SiblingNegotiation ~ FoodTreatment*SexParent + offset(log(BroodSize)) + (1|Nest), ziformula = ~ FoodTreatment + SexParent,  data=Owls , family = nbinom1)
 
 res <- simulateResiduals(m4, plot = T)
@@ -610,7 +608,7 @@ testZeroInflation(res)
 
 This looks a lot better. Trying a slightly different model specification 
 
-```{r}
+```{r, eval= F}
 m5 <- glmmTMB(SiblingNegotiation ~ FoodTreatment*SexParent + offset(log(BroodSize)) + (1|Nest), dispformula = ~ FoodTreatment , ziformula = ~ FoodTreatment + SexParent,  data=Owls , family = nbinom1)
 
 res <- simulateResiduals(m4, plot = T)