reservr0.0.3 package

Fit Distributions and Neural Networks to Censored and Truncated Data

dist_gamma

Gamma distribution

dist_pareto

Pareto Distribution

dist_poisson

Poisson Distribution

dist_translate

Tranlsated distribution

dist_beta

Beta Distribution

dist_binomial

Binomial Distribution

as_params

Convert TensorFlow tensors to distribution parameters recursively

blended_transition

Transition functions for blended distributions

callback_adaptive_lr

Keras Callback for adaptive learning rate with weight restoration

callback_debug_dist_gradients

Callback to monitor likelihood gradient components

dist_bdegp

Construct a BDEGP-Family

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_genpareto

Generalized Pareto Distribution

dist_lognormal

Log Normal distribution

dist_mixture

Mixture distribution

dist_negbinomial

Negative binomial Distribution

dist_normal

Normal distribution

dist_trunc

Truncated distribution

dist_uniform

Uniform distribution

dist_weibull

Weibull Distribution

Distribution

Base class for Distributions

fit_blended

Fit a Blended mixture using an ECME-Algorithm

fit_dist_start

Find starting values for distribution parameters

Pareto

The Pareto Distribution

fit_dist

Fit a general distribution to observations

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

flatten_params

Flatten / Inflate parameter lists / vectors

GenPareto

The Generalized Pareto Distribution (GPD)

integrate_gk

Adaptive Gauss-Kronrod Quadrature for multiple limits

interval-operations

Convex union and intersection of intervals

interval

Intervals

is.Distribution

Test if object is a Distribution

k_matrix

Cast to a TensorFlow matrix

weighted_quantile

Compute weighted quantiles

plot_distributions

Plot several distributions

predict.reservr_keras_model

Predict individual distribution parameters

prob_report

Determine probability of reporting under a Poisson arrival Process

quantile.Distribution

Quantiles of Distributions

reexports

Objects exported from other packages

softmax

Soft-Max function

weighted_tabulate

Compute weighted tabulations

tf_compile_model

Compile a Keras model for truncated data under dist

tf_initialise_model

Initialise model weights to a global parameter fit

trunc_obs

Define a set of truncated observations

truncate_claims

Truncate claims data subject to reporting delay

weighted_moments

Compute weighted moments

Define distribution families and fit them to interval-censored and interval-truncated data, where the truncation bounds may depend on the individual observation. The defined distributions feature density, probability, sampling and fitting methods as well as efficient implementations of the log-density log f(x) and log-probability log P(x0 <= X <= x1) for use in 'TensorFlow' neural networks via the 'tensorflow' package. Allows training parametric neural networks on interval-censored and interval-truncated data with flexible parameterization. Applications include Claims Development in Non-Life Insurance, e.g. modelling reporting delay distributions from incomplete data, see Bücher, Rosenstock (2022) <doi:10.1007/s13385-022-00314-4>.

  • Maintainer: Alexander Rosenstock
  • License: GPL
  • Last published: 2024-06-24