A method to plot and/or return the predicted tail factors for a specific attachment point.
The object from which to plot the predicted tail factors and return tail factors for all attachment points.
An integer value specifying the attachment point for the tail. Must be at least 1. See Details for more info.
A logical value. If TRUE
, observed values are substituted for predicted values whenever possible in the calculation. If FALSE
, only predicted values are used.
A logical value or NA
. See Details for more information.
An integer value must be at least 1 default value is attachment
. A call to tailFactor
returns (invisibly) a matrix of tail factors through this value.
A logical value. If TRUE
, the plot is generated and the statistics are returned; otherwise only the statistics are returned.
Either a range of years (for example c(1995, 2006)) or one of the keywords “all” or “fullyObs”.
Mainly called for the side effect of plotting. Also returns tail factors for all attachment points through finalAttachment
. Returned invisibly.
The tail factor is the ratio of the estimated ultimate loss to cumulative loss at some point in development time. This is a method to allow for the retrieval and illustration of the tail factor by exposure year.
Because the model is Bayesian, each tail factor comes as a distribution. To ease graphical interpretation, only the median for each factor is plotted/returned.
See for more details tailFactor
.