This function generates proto_ipm objects from
Padrino Database tables.
pdb_make_proto_ipm(pdb, ipm_id = NULL, det_stoch = "det", kern_param = "kern")A pdb object.
Optionally, one or more ipm_id's to build. If empty,
all models contained in the pdb object will be processed into
proto_ipm's.
A vector containing either "det" or "stoch".
This determines whether we want to construct a deterministic or stochastic
model. Default is "det". See details
If det_stoch = "stoch", then whether or not to construct
a kernel resampled model, or a parameter resampled model. See details.
A list containing one or more proto_ipms. Names of the list
will correspond to ipm_ids.
proto_ipm objects contain all of the information needed
to implement an IPM, but stop short of actually generating kernels. These
are intermediate building blocks that can be modified before creating a full
IPM so that things like perturbation analysis are a bit more straightforward.
When requesting many models, the det_stoch and kern_param parameters
can also be vectors. These are matched with ipm_id by position. If the
lengths of det_stoch and kern_param do not match the length
ipm_id, they will be recycled until they do.
For stochastic models, there is sometimes the option of building either a kernel-resampled or a parameter resampled model. A kernel resampled model uses some point estimate for time and/or space varying parameters to generate kernels for each year/site/grouping factor. Parameter resampled models sample parameters from distributions. Padrino stores this information for some models when it is available in the literature, and tries to fail informatively when these distributions aren't available in the database.
For more info on kern_param definitions:
Metcalf et al. (2015). Statistial modeling of annual variation for inference on stochastic population dynamics using Integral Projection Models. Methods in Ecology and Evolution. DOI: 10.1111/2041-210X.12405