lqr (version 5.2)

SKD: Skew Family Distributions

Description

Density, distribution function, quantile function and random generation for a Skew Family Distribution useful for quantile regression. This family of distribution includes skewed versions of the Normal, Student's t, Laplace, Slash and Contaminated Normal distribution, all with location parameter equal to mu, scale parameter sigma and skewness parameter p.

Usage

dSKD(y, mu = 0, sigma = 1, p = 0.5, dist = "normal", nu = "", gamma = "")
pSKD(q, mu = 0, sigma = 1, p = 0.5, dist = "normal", nu = "", gamma = "",
lower.tail = TRUE)
qSKD(prob, mu = 0, sigma = 1, p = 0.5, dist = "normal", nu = "", gamma = "",
lower.tail = TRUE)
rSKD(n, mu = 0, sigma = 1, p = 0.5, dist = "normal", nu = "", gamma = "")

Value

dSKD gives the density, pSKD gives the distribution function, qSKD gives the quantile function, and rSKD generates a random sample.

The length of the result is determined by n for rSKD, and is the maximum of the lengths of the numerical arguments for the other functions dSKD, pSKD and qSKD.

Arguments

y,q

vector of quantiles.

prob

vector of probabilities.

n

number of observations.

mu

location parameter.

sigma

scale parameter.

p

skewness parameter.

dist

represents the distribution to be used for the error term. The values are normal for Normal distribution, t for Student's t distribution, laplace for Laplace distribution, slash for Slash distribution and cont for the Contaminated normal distribution.

nu

It represents the degrees of freedom when dist = t. For the Slash distribution (dist = slash) it is a shape parameter \(\nu>0\). For the Contaminated Normal distribution, \(\nu\) is the parameter that represents the percentage of outliers. When is not provided, we use the MLE.

gamma

It represents a scale factor for the contaminated normal distribution. When is not provided, we use the MLE.

lower.tail

logical; if TRUE (default), probabilities are P[X \(\le\) x] otherwise, P[X > x].

Author

Christian E. Galarza <>, Luis Benites <> and Victor H. Lachos <>

Maintainer: Christian E. Galarza <>

Details

If mu, sigma, p or dist are not specified they assume the default values of 0, 1, 0.5 and normal, respectively, belonging to the Symmetric Standard Normal Distribution denoted by \(SKN(0,1,0.5)\).

The scale parameter sigma must be positive and non zero. The skew parameter p must be between zero and one (0<p<1).

This family of distributions generalize the skew distributions in Wichitaksorn et.al. (2014) as an scale mixture of skew normal distribution. Also the Three-Parameter Asymmetric Laplace Distribution defined in Koenker and Machado (1999) is a special case.

References

Galarza, C., Lachos, V. H., Cabral, C. R. B., & Castro, C. L. (2017). Robust quantile regression using a generalized class of skewed distributions. Stat, 6(1), 113-130.

Wichitaksorn, N., Choy, S. T., & Gerlach, R. (2014). A generalized class of skew distributions and associated robust quantile regression models. Canadian Journal of Statistics, 42(4), 579-596.

See Also

lqr,ais

Examples

# \donttest{
## Let's plot (Normal Vs. Student-t's with 4 df)
##Density
sseq = seq(15,65,length.out = 1000)
dens = dSKD(y=sseq,mu=50,sigma=3,p=0.75)
plot(sseq,dens,type="l",lwd=2,col="red",xlab="x",ylab="f(x)", main="Normal Vs. t(4) densities")
dens2 = dSKD(y=sseq,mu=50,sigma=3,p=0.75,dist="t",nu=4)
lines(sseq,dens2,type="l",lwd=2,col="blue",lty=2)

## Distribution Function
df = pSKD(q=sseq,mu=50,sigma=3,p=0.75,dist = "laplace")
plot(sseq,df,type="l",lwd=2,col="blue",xlab="x",ylab="F(x)", main="Laplace Distribution function")
abline(h=1,lty=2)

##Inverse Distribution Function
prob = seq(0.001,0.999,length.out = 1000)
idf = qSKD(prob=prob,mu=50,sigma=3,p=0.25,dist="cont",nu=0.3,gamma=0.1) # 1 min appox
plot(prob,idf,type="l",lwd=2,col="gray30",xlab="x",ylab=expression(F^{-1}~(x)))
title(main="Skew Cont. Normal Inverse Distribution function")
abline(v=c(0,1),lty=2)

#Random Sample Histogram
sample = rSKD(n=20000,mu=50,sigma=3,p=0.2,dist="slash",nu=3)
seqq2 = seq(25,100,length.out = 1000)
dens3 = dSKD(y=seqq2,mu=50,sigma=3,p=0.2,dist="slash",nu=3)
hist(sample,breaks = 70,freq = FALSE,ylim=c(0,1.05*max(dens3,na.rm = TRUE)),main="")
title(main="Histogram and True density")
lines(seqq2,dens3,col="blue",lwd=2)
# }