BLRPM.OF {BLRPM}R Documentation

BLRPM objective function for parameter estimation


BLRPM.OF is the objective function used for parameter estimation of the BLRPM parameters. Given a set of BLRPM parameters par this function calculates a set of model statistics at given accumulation time steps acc.vals. These model statistics are compared with given time series statistics stats in the objective function. The user is able to define weights for each statistic (has to be the same length as statistics input vector). Option for debugging is given. A scale parameter defines a criterium for which different kinds of model statistics are calculated. This criterium is mainly based on the timescale difference between storm duration parameter gamma and cell duration parameter eta. If use.log is true, the objective function needs logarithmic input parameters. The value of OF defines the kind of objective function to be used: 1= quadratic 2= quadratic extended 3= absolute 4= absolute extended.


BLRPM.OF(par, stats, acc.vals = c(1, 3, 12, 24), weights = rep(1,
  length(stats)), debug = FALSE, scale = 1, use.log = TRUE, OF = 2)



vector specifying the five model parameters (lambda,gamma,beta,eta,mux) at which the objective function is to be calculated


vector specifying the time series statistics to which the model is compared to


vector specifying the accumulation time steps at which the model statistics are calculated


vector specifying the weight of each statistic in the objective function. Note: has to have the same length as stats


logical defining if debuggging of function has to be done


value spacifying the scale factor for comparisson between duration parameters gamma and eta


logical defining if input parameters are logarithmic


value specifying the type of objective function. 1: quadratic, 2: quad extended, 3: absolute, 4: abs extended


Z returns value of objective function for input parameters and input statistics


Christoph Ritschel

[Package BLRPM version 1.0 Index]