allow intercept in process formula if trend_map supplied

nicholasjclark · Jul 2, 2024 · 87c6077 · 87c6077
1 parent 748a46a
commit 87c6077
Show file tree

Hide file tree

Showing 184 changed files with 1,093 additions and 1,437 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,7 +1,7 @@
 Package: mvgam
 Title: Multivariate (Dynamic) Generalized Additive Models
-Version: 1.1.2
-Date: 2024-06-28
+Version: 1.1.3
+Date: 2024-07-02
 Authors@R: person("Nicholas J", "Clark", , "nicholas.j.clark1214@gmail.com", role = c("aut", "cre"), comment = c(ORCID = "0000-0001-7131-3301"))
 Description: Fit Bayesian Dynamic Generalized Additive Models to sets of time series. Users can build dynamic nonlinear State-Space models that can incorporate semiparametric effects in observation and process components, using a wide range of observation families. Estimation is performed using Markov Chain Monte Carlo with Hamiltonian Monte Carlo in the software 'Stan'. References: Clark & Wells (2022) <doi:10.1111/2041-210X.13974>.
 URL: https://github.com/nicholasjclark/mvgam, https://nicholasjclark.github.io/mvgam/

diff --git a/NEWS.md b/NEWS.md
@@ -1,4 +1,9 @@
+# mvgam 1.1.3
+## New functionalities
+* Added option for process model intercept parameter to be included when a `trend_map` is supplied. This breaks the assumption that the process has to be zero-centred, adding flexibility but also potentially inducing nonidentifiabilities with respect to any observation model intercepts. Thoughtful priors are a must for these models
+
 # mvgam 1.1.2
+## New functionalities
 * Added options for silencing some of the 'Stan' compiler and modeling messages using the `silent` argument in `mvgam()`
 * Moved a number of packages from 'Depends' to 'Imports' for simpler package loading and fewer potential masking conflicts
 * Improved efficiency of the model initialisation by tweaking parameters of the underlying 'mgcv' `gam` object's convergence criteria, resulting in much faster model setups
@@ -8,9 +13,12 @@ which speeds up mixing most of the time
 * Updated support for multithreading so that all observation families (apart from `nmix()`) can now be modeled with multiple threads
 * Changed default priors on autoregressive coefficients (AR1, AR2, AR3) to enforce
 stationarity, which is a much more sensible prior in the majority of contexts
+
+## Bug fixes
 * Fixed a small bug that prevented `conditional_effects.mvgam()` from handling effects with three-way interactions
 
 # mvgam 1.1.1
+## New functionalities
 * Changed indexing of an internal c++ function after Prof Brian Ripley’s   
   email: Dear maintainer, Please see the problems shown on 
   https://cran.r-project.org/web/checks/check_results_mvgam.html. Please correct   before 2024-05-22 to safely retain your package on CRAN. The CRAN Team

diff --git a/R/get_mvgam_priors.R b/R/get_mvgam_priors.R
@@ -248,8 +248,11 @@ get_mvgam_priors = function(formula,
                                         fixed = TRUE),
                                    collapse = " "))
 
-    # Drop any intercept from the formula if this is not an N-mixture model
-    if(family_char != 'nmix'){
+    # Drop any intercept from the formula if this is not an N-mixture model or if a
+    # trend_map was not originally supplied
+    if(family_char == 'nmix') drop_trend_int <- FALSE else drop_trend_int <- TRUE
+    if(!missing(trend_map)) drop_trend_int <- FALSE
+    if(drop_trend_int){
       if(attr(terms(trend_formula), 'intercept') == 1){
         trend_formula <- update(trend_formula, trend_y  ~ . -1)
       } else {

diff --git a/R/mvgam.R b/R/mvgam.R
@@ -27,7 +27,8 @@
 #'(i.e. by allowing effects to vary across process models, even when some time series share the same underlying
 #'process model). This feature is only currently available for `RW()`, `AR()` and `VAR()` trend models.
 #'In `nmix()` family models, the `trend_formula` is used to set up a linear predictor for the underlying
-#'latent abundance
+#'latent abundance. If this is supplied but you do not also supply a `trend_map`,
+#'the intercept parameter in this formula will automatically be surpressed to maintain identifiability. In other words, the latent process is assumed to be zero-centred
 #'@param knots An optional \code{list} containing user specified knot values to be used for basis construction.
 #'For most bases the user simply supplies the knots to be used, which must match up with the k value supplied
 #'(note that the number of knots is not always just `k`). Different terms can use different numbers of knots,
@@ -123,8 +124,9 @@
 #'`use_lv = TRUE` and using the mapping to set up the shared trends. Needs to have column names
 #'`series` and `trend`, with integer values in the `trend` column to state which trend each series
 #'should depend on. The `series` column should have a single unique entry for each series in the
-#'data (names should perfectly match factor levels of the `series` variable in `data`). See examples
-#'for details
+#'data (names should perfectly match factor levels of the `series` variable in `data`). Note that
+#'if this is supplied, the intercept parameter in the process model will NOT be automatically suppressed.
+#'See examples for details
 #'@param drift \code{logical} estimate a drift parameter in the latent trend components. Useful if the latent
 #'trend is expected to broadly follow a non-zero slope. Only available for
 #'`RW()` and `AR()` trend models. Note that if the latent trend is more or less stationary,

diff --git a/docs/news/index.html b/docs/news/index.html
diff --git a/docs/reference/Rplot001.png b/docs/reference/Rplot001.png
diff --git a/docs/reference/Rplot002.png b/docs/reference/Rplot002.png
diff --git a/docs/reference/Rplot003.png b/docs/reference/Rplot003.png
diff --git a/docs/reference/Rplot004.png b/docs/reference/Rplot004.png
diff --git a/docs/reference/Rplot005.png b/docs/reference/Rplot005.png
diff --git a/docs/reference/Rplot006.png b/docs/reference/Rplot006.png
diff --git a/docs/reference/Rplot007.png b/docs/reference/Rplot007.png
diff --git a/docs/reference/Rplot008.png b/docs/reference/Rplot008.png
diff --git a/docs/reference/Rplot009.png b/docs/reference/Rplot009.png
diff --git a/docs/reference/Rplot010.png b/docs/reference/Rplot010.png
diff --git a/docs/reference/Rplot011.png b/docs/reference/Rplot011.png
diff --git a/docs/reference/Rplot012.png b/docs/reference/Rplot012.png
diff --git a/docs/reference/Rplot013.png b/docs/reference/Rplot013.png
diff --git a/docs/reference/Rplot014.png b/docs/reference/Rplot014.png
diff --git a/docs/reference/Rplot015.png b/docs/reference/Rplot015.png
diff --git a/docs/reference/Rplot016.png b/docs/reference/Rplot016.png
diff --git a/docs/reference/Rplot018.png b/docs/reference/Rplot018.png
diff --git a/docs/reference/Rplot019.png b/docs/reference/Rplot019.png
diff --git a/docs/reference/Rplot020.png b/docs/reference/Rplot020.png
diff --git a/docs/reference/Rplot021.png b/docs/reference/Rplot021.png
diff --git a/docs/reference/Rplot022.png b/docs/reference/Rplot022.png
diff --git a/docs/reference/Rplot023.png b/docs/reference/Rplot023.png
diff --git a/docs/reference/Rplot024.png b/docs/reference/Rplot024.png
diff --git a/docs/reference/Rplot025.png b/docs/reference/Rplot025.png
diff --git a/docs/reference/Rplot026.png b/docs/reference/Rplot026.png
diff --git a/docs/reference/Rplot027.png b/docs/reference/Rplot027.png
diff --git a/docs/reference/Rplot028.png b/docs/reference/Rplot028.png
diff --git a/docs/reference/conditional_effects.mvgam-1.png b/docs/reference/conditional_effects.mvgam-1.png
diff --git a/docs/reference/conditional_effects.mvgam-10.png b/docs/reference/conditional_effects.mvgam-10.png
diff --git a/docs/reference/conditional_effects.mvgam-11.png b/docs/reference/conditional_effects.mvgam-11.png
diff --git a/docs/reference/conditional_effects.mvgam-12.png b/docs/reference/conditional_effects.mvgam-12.png
diff --git a/docs/reference/conditional_effects.mvgam-2.png b/docs/reference/conditional_effects.mvgam-2.png
diff --git a/docs/reference/conditional_effects.mvgam-3.png b/docs/reference/conditional_effects.mvgam-3.png
diff --git a/docs/reference/conditional_effects.mvgam-4.png b/docs/reference/conditional_effects.mvgam-4.png
diff --git a/docs/reference/conditional_effects.mvgam-5.png b/docs/reference/conditional_effects.mvgam-5.png
diff --git a/docs/reference/conditional_effects.mvgam-6.png b/docs/reference/conditional_effects.mvgam-6.png
diff --git a/docs/reference/conditional_effects.mvgam-7.png b/docs/reference/conditional_effects.mvgam-7.png
diff --git a/docs/reference/conditional_effects.mvgam-8.png b/docs/reference/conditional_effects.mvgam-8.png
diff --git a/docs/reference/conditional_effects.mvgam-9.png b/docs/reference/conditional_effects.mvgam-9.png