Function reference • reservr

Distributions

Specifying families of distributions to use for fitting, simulation or prediction.

Distribution: Base class for Distributions

is.Distribution(): Test if object is a Distribution

dist_bdegp(): Construct a BDEGP-Family

dist_beta(): Beta Distribution

dist_binomial(): Binomial Distribution

dist_blended(): Blended distribution

dist_dirac(): Dirac (degenerate point) Distribution

dist_discrete(): Discrete Distribution

dist_empirical(): Empirical distribution

dist_erlangmix(): Erlang Mixture distribution

dist_exponential(): Exponential distribution

dist_gamma(): Gamma distribution

dist_genpareto() dist_genpareto1(): Generalized Pareto Distribution

dist_lognormal(): Log Normal distribution

dist_mixture(): Mixture distribution

dist_negbinomial(): Negative binomial Distribution

dist_normal(): Normal distribution

dist_pareto(): Pareto Distribution

dist_poisson(): Poisson Distribution

dist_translate(): Tranlsated distribution

dist_trunc(): Truncated distribution

dist_uniform(): Uniform distribution

dist_weibull(): Weibull Distribution

plot_distributions(): Plot several distributions

quantile(<Distribution>): Quantiles of Distributions

Work with truncated data

Specify and work with interval truncated and interval censored data

trunc_obs() as_trunc_obs() truncate_obs() repdel_obs(): Define a set of truncated observations

truncate_claims(): Truncate claims data subject to reporting delay

Fit Models on Data

Fit models to truncated data using maximum likelihood or EM-Algorithms, depending on the selected family.

fit_blended(): Fit a Blended mixture using an ECME-Algorithm

fit_dist() fit_dist_direct() fit(<Distribution>): Fit a general distribution to observations

fit_dist_start(): Find starting values for distribution parameters

fit_erlang_mixture(): Fit an Erlang mixture using an ECME-Algorithm

fit_mixture(): Fit a generic mixture using an ECME-Algorithm

fit(<reservr_keras_model>): Fit a neural network based distribution model to data

Extract predictions from fitted models

Methods for adding predictions to new data.

prob_report(): Determine probability of reporting under a Poisson arrival Process

predict(<reservr_keras_model>): Predict individual distribution parameters

Work with TensorFlow

Methods for specifying, compiling and creating input for TensorFlow models using the tensorflow package.

tf_compile_model(): Compile a Keras model for truncated data under dist

tf_initialise_model(): Initialise model weights to a global parameter fit

k_matrix(): Cast to a TensorFlow matrix

callback_adaptive_lr(): Keras Callback for adaptive learning rate with weight restoration

callback_debug_dist_gradients(): Callback to monitor likelihood gradient components

as_params(): Convert TensorFlow tensors to distribution parameters recursively

Miscellaneous

Miscellaneous functions

blended_transition() blended_transition_inv(): Transition functions for blended distributions

flatten_params() flatten_params_matrix() flatten_bounds() inflate_params(): Flatten / Inflate parameter lists / vectors

rgpd() dgpd() pgpd() qgpd(): The Generalized Pareto Distribution (GPD)

rpareto() dpareto() ppareto() qpareto(): The Pareto Distribution

integrate_gk(): Adaptive Gauss-Kronrod Quadrature for multiple limits

interval() is.Interval(): Intervals

interval_union() interval_intersection(): Convex union and intersection of intervals

softmax() dsoftmax(): Soft-Max function

weighted_moments(): Compute weighted moments

weighted_quantile() weighted_median(): Compute weighted quantiles

weighted_tabulate(): Compute weighted tabulations