Fit of univariate distribution by matching quantiles for non censored data.
qmedist(data, distr, probs, start = NULL, fix.arg = NULL, qtype = 7,
optim.method = "default", lower = -Inf, upper = Inf,
custom.optim = NULL, weights = NULL, silent = TRUE, gradient = NULL,
checkstartfix=FALSE, …)A numeric vector for non censored data.
A character string "name" naming a distribution for which the corresponding
quantile function
qname and the corresponding density distribution dname must be classically defined.
A numeric vector of the probabilities for which the quantile matching is done. The length of this vector must be equal to the number of parameters to estimate.
A named list giving the initial values of parameters of the named distribution
or a function of data computing initial values and returning a named list.
This argument may be omitted (default) for some distributions for which reasonable
starting values are computed (see the 'details' section of mledist).
An optional named list giving the values of fixed parameters of the named distribution or a function of data computing (fixed) parameter values and returning a named list. Parameters with fixed value are thus NOT estimated.
The quantile type used by the R quantile function to
compute the empirical quantiles, (default 7 corresponds to the default quantile method in R).
"default" or optimization method to pass to optim.
Left bounds on the parameters for the "L-BFGS-B" method (see optim).
Right bounds on the parameters for the "L-BFGS-B" method (see optim).
a function carrying the optimization.
an optional vector of weights to be used in the fitting process.
Should be NULL or a numeric vector with strictly positive integers
(typically the number of occurences of each observation). If non-NULL,
weighted QME is used, otherwise ordinary QME.
A logical to remove or show warnings when bootstraping.
A function to return the gradient of the squared difference for the "BFGS", "CG"
and "L-BFGS-B" methods. If it is NULL, a finite-difference approximation will be used,
see details.
A logical to test starting and fixed values. Do not change it.
further arguments passed to the optim,
constrOptim or custom.optim function.
qmedist returns a list with following components,
the parameter estimates.
an integer code for the convergence of optim defined as below
or defined by the user in the user-supplied optimization function.
0 indicates successful convergence.
1 indicates that the iteration limit of optim has been reached.
10 indicates degeneracy of the Nealder-Mead simplex.
100 indicates that optim encountered an internal error.
the minimal value reached for the criterion to minimize.
a symmetric matrix computed by optim as an estimate of the Hessian
at the solution found or computed in the user-supplied optimization function.
the name of the optimization function used for maximum likelihood.
when optim is used, the name of the
algorithm used, the field method of the custom.optim function
otherwise.
the named list giving the values of parameters of the named distribution
that must kept fixed rather than estimated by maximum likelihood or NULL
if there are no such parameters.
the function used to set the value of fix.arg or NULL.
the vector of weigths used in the estimation process or NULL.
A two-element integer vector giving the number of calls
to the log-likelihood function and its gradient respectively.
This excludes those calls needed to compute the Hessian, if requested,
and any calls to log-likelihood function to compute a finite-difference
approximation to the gradient. counts is returned by optim
or the user-supplied function or set to NULL.
A character string giving any additional information
returned by the optimizer, or NULL. To understand exactly the message,
see the source code.
the log-likelihood value.
the probability vector on which quantiles are matched.
The qmedist function carries out the quantile matching numerically, by minimization of the
sum of squared differences between observed and theoretical quantiles.
Note that for discrete distribution, the sum of squared differences is a step function and
consequently, the optimum is not unique, see the FAQ.
The optimization process is the same as mledist, see the 'details' section
of that function.
Optionally, a vector of weights can be used in the fitting process.
By default (when weigths=NULL), ordinary QME is carried out, otherwise
the specified weights are used to compute weighted quantiles used in the squared differences.
Weigthed quantiles are computed by wtd.quantile from the Hmisc package.
It is not yet possible to take into account weighths in functions plotdist,
plotdistcens, plot.fitdist, plot.fitdistcens, cdfcomp,
cdfcompcens, denscomp, ppcomp, qqcomp, gofstat
and descdist
(developments planned in the future).
This function is not intended to be called directly but is internally called in
fitdist and bootdist.
Klugman SA, Panjer HH and Willmot GE (2012), Loss Models: From Data to Decissions, 4th edition. Wiley Series in Statistics for Finance, Business and Economics, p. 253.
Delignette-Muller ML and Dutang C (2015), fitdistrplus: An R Package for Fitting Distributions. Journal of Statistical Software, 64(4), 1-34.
mmedist, mledist, mgedist,
fitdist for other estimation methods and
quantile for empirical quantile estimation in R.
# NOT RUN {
# (1) basic fit of a normal distribution
#
set.seed(1234)
x1 <- rnorm(n=100)
qmedist(x1, "norm", probs=c(1/3, 2/3))
# (2) defining your own distribution functions, here for the Gumbel
# distribution for other distributions, see the CRAN task view dedicated
# to probability distributions
dgumbel <- function(x, a, b) 1/b*exp((a-x)/b)*exp(-exp((a-x)/b))
qgumbel <- function(p, a, b) a - b*log(-log(p))
qmedist(x1, "gumbel", probs=c(1/3, 2/3), start=list(a=10,b=5))
# (3) fit a discrete distribution (Poisson)
#
set.seed(1234)
x2 <- rpois(n=30,lambda = 2)
qmedist(x2, "pois", probs=1/2)
# (4) fit a finite-support distribution (beta)
#
set.seed(1234)
x3 <- rbeta(n=100,shape1=5, shape2=10)
qmedist(x3, "beta", probs=c(1/3, 2/3))
# (5) fit frequency distributions on USArrests dataset.
#
x4 <- USArrests$Assault
qmedist(x4, "pois", probs=1/2)
qmedist(x4, "nbinom", probs=c(1/3, 2/3))
# }
Run the code above in your browser using DataLab