TODO 0.3.0

[Kertoo]

3. Plan for singleRcapture version 0.3.0

Refactor code to make it easier to implement non standard approaches:

• Subclasses for singleR (currently only one for standard implemented frequentist count modelling later add class for bayesian and mixed effects)
• Change methods to adjust for these subclasses
• Make it so that family type objects fully handle data (i.e. with zero one truncated or chao/zelterman)
• Implement ratio regression

Further diagnostic information:

• Null deviance
• deviance residuals (for models that do require a paper with theory)
• ...

Fitting:

• Bias Reduction compatibility with brglm2
• Robust zero-truncated Poisson robust inference for ztp and ztop #5
• Robust zero-truncated Geometric
• EM algorithm
• Estimation by full likelihood i.e. not observed likelihood
• Link functions in family functions
• Additional measures for intercept only models to make them faster

Quality of life improvements:

• Prior weights as population counts
• Class for multiple models
• Multithreading (or possibly julia backend if it will be easier) in bootstrap and multiplication of $\boldsymbol{W}_{k}$ matrixes with pseudo-residuals and simulation of information information matrixes in negative binomials
• Update the simulate.singleR method and create simulate.singleRfamily, update.singleR methods.
• Method for variable selection.
• ...

Compatibility:

• Method pseudoR2 funcions for singleR class

Models:

• truncated binomial models (inflated hurdle etc)
• multiple chao
• one-inflated chao

Moved to singleRcaptureExtra package:

• Method for pop size estimation for vglm
• Method for pop size estimation for zerotrunc