NEWS.md

In ulrichriegel/Pareto: The Pareto, Piecewise Pareto and Generalized Pareto Distribution

• Pareto 2.4.5

• Minor bug fix in automated tests

Pareto 2.4.3

• Minor bug corrected in PiecewisePareto_ML_Estimator_Alpha

Pareto 2.4.2

• Added functionality for Pareto and GenPareto in Fit_References

Pareto 2.4.0

• Improved functionality for maximum likelihood estimation
  • Possibility to use reporting thresholds
  • Allow to consider censored data
  • Improved performance
• Added distributions in function Local_Pareto_Alpha:
  • Pareto distribution
  • Generalized Pareto distribution
  • Piecewise Pareto distribution
• Improved handling of inputs of length zero in vectorized functions

Pareto 2.3.0

• Vectorization of the following functions:
  • Pareto_Layer_Mean
  • Pareto_Layer_Var
  • Pareto_Layer_SM
  • Pareto_Extrapolation
  • Pareto_Find_Alpha_btw_Layers
  • Pareto_Find_Alpha_btw_FQ_Layer
  • Pareto_Find_Alpha_btw_FQs
  • PiecewisePareto_Layer_Mean (only parameters Cover and AttachmentPoint)
  • PiecewisePareto_Layer_SM (only parameters Cover and AttachmentPoint)
  • PiecewisePareto_Layer_Var (only parameters Cover and AttachmentPoint)
  • pPareto
  • dPareto
  • qPareto
  • pGenPareto
  • dGenPareto
  • qGenPareto
  • GenPareto_Layer_Mean
  • GenPareto_Layer_Var
  • GenPareto_Layer_SM

Pareto 2.2.2

• Added function Fit_PML_Curve which fits a PPP_Model to a PML curve..

Pareto 2.2.1

• Added the option to use weights in Pareto_ML_Estimator_Alpha, PiecewisePareto_ML_Estimator_Alpha and GenPareto_ML_Estimator_Alpha.

Pareto 2.2.0

• Added function Fit_References for the piecewise Pareto distribution. This function fits a PPP model to the expected losses of given reference layers and excess frequencies
• It is now possible to have layers with an expected loss of zero in PiecewisePareto_Match_Layer_Losses
• Improved handling of Frequencies and TotalLoss_Frequencies in PiecewisePareto_Match_Layer_Losses

Pareto 2.1.0

• Added functions for the generalized Pareto distribution
• Added the class PGP_Model. PGP stands for Panjer & Generalized Pareto. A PGP_Model object contains the information to specify a collective model with a Panjer distributed claim count and a generalized Pareto distributed severity
• The following functions have been replaced by generics for PPP_Models and PGP_Models:
  • PPP_Model_Exp_Layer_Loss has been replaced by Layer_Mean
  • PPP_Model_Layer_Var has been replaced by Layer_Var
  • PPP_Model_Layer_Sd has been replaced by Layer_Sd
  • PPP_Model_Excess_Frequency has been replaced by Excess_Frequency
  • PPP_Model_Simulate has been replaced by Simulate_Losses

Pareto 2.0.0

• PiecewisePareto_Match_Layer_Losses now returns a PPP_Model object. PPP stands for Panjer & Piecewise Pareto. The Panjer class contains the Poisson, the Negative Binomial and the Binomial distribution. A PPP_Model object contains the information required to specify a collective model with a Panjer distributed claim count and a Piecewise Pareto distributed severity.
• The package provides additional functions for PPP_Model objects:
  • PPP_Model_Exp_Layer_Loss: Calculates the expected loss of a reinsurance layer for a PPP_Model
  • PPP_Model_Layer_Var: Calculates the variance of the loss in a reinsurance layer for a PPP_Model
  • PPP_Model_Layer_Sd: Calculates the standard deviation of the loss in a reinsurance layer for a PPP_Model
  • PPP_Model_Excess_Frequency: Calculates the expected frequency in excess of a threshold for a PPP_Model
  • PPP_Model_Simulate: Simulates losses of a PPP_Model

Pareto 1.1.5

• PiecewisePareto_Match_Layer_Losses now also works for only one layer
• Improved error handling in PiecewisePareto_Match_Layer_Losses

Pareto 1.1.3

• Added maximum likelihood estimation of the alphas of a piecewise Pareto distribution.
• Allow for a different reporting threshold for each loss in Pareto_ML_Estimator_Alpha and in rPareto.
• Improved fitting algorithm in Pareto_ML_Estimator_Alpha.
• Better error handling in in Pareto_Find_Alpha_btw_FQ_Layer.

Pareto 1.1.0

Stable version.

ulrichriegel/Pareto documentation built on April 21, 2023, 8:39 p.m.