首页 > 学术百科

r语言quantmond_R中的关于极值理论的包

r语⾔quantmond_R中的关于极值理论的包第⼀部分：包evir

⼀、探索性函数：

library(evir)

data(danish)

findthresh(danish, 50)

寻阀值，例⼦中寻出来的阀值使得超越它的为50个数。

data(danish)

emplot(danish) #经验分布函数，如果得到的结果是直线那么符合帕累托分布。

data(bmw)

exindex(bmw, 100) #估计极值指标

data(danish)

hill(danish) #Hill 图基于帕累托分布

data(danish)

meplot(danish) #简单均值剩余图

如果图形是向上直线，说明有厚尾，如果直线有正的斜率，说明服从帕累托分布。图

形如果是向下的直线，说明薄尾，如果直线0斜率，服从指数分布。

data(danish)

records(danish) #极值记录函数

⼆、分布拟合函数

data(bmw)

out

对甲苯磺酸甲酯

data(nidd.annual)

out

拟合gumbel分布。

data(danish)

out

data(danish)

shape(danish) #估计GPD参数图

data(danish)

quant(danish, 0.999) #GPD⾼分位数尾估计图

tp=tailplot(out) #Plot Tail Estimate From GPD Model

gpd.q(tp, 0.999) # Add Quantile Estimates to plot.gpd

gpd.sfall(tp, 0.999) # Add Expected Shortfall Estimates to a GPD Plot data(danish)

out

拟合POT模型

data(bmw)

data(siemens)

out

interpret.gpdbiv(out, 0.05, 0.05) #解释⼆元gpd模型

plot(out) #其中有5张图可以画

三、风险损失估计函数

网上大学

data(danish)

out

riskmeasures(out, c(0.999, 0.9999))

第⼆部分：包evd

⼀、探索型函数

par(mfrow = c(1,2))

smdat1

smdat2

chiplot(smdat1) #针对⼆元极值度量依赖度依赖为1，否则接近于0.

chiplot(smdat2) # 独⽴为0

data(portpirie)

tlim

tcplot(portpirie, tlim) #阀值选择

tcplot(portpirie, tlim, nt = 100, lwd = 3, type = "l")

tcplot(portpirie, tlim, model = "pp")

data(portpirie)

clusters(portpirie, 4.2, 3) #甄别超限簇

clusters(portpirie, 4.2, 3, cmax = TRUE) #cmax=F

clusters(portpirie, 4.2, 3, 3.8, plot = TRUE)

clusters(portpirie, 4.2, 3, 3.8, plot = TRUE, lvals = FALSE)意思自治

data(portpirie)

exi(portpirie, 4.2, r = 3, ulow = 3.8) #估计极值指标

tvu

exi(portpirie, tvu, r = 1)

exi(portpirie, tvu, r = 0)

sdat

tlim

exiplot(sdat, tlim) #画极值指标图

exiplot(sdat, tlim, r = 4, add = TRUE, lty = 2)

exiplot(sdat, tlim, r = 0, add = TRUE, lty = 4)

data(portpirie)

mrlplot(portpirie) #经验平均剩余寿命图

⼆、分布估计函数

bvdata

abvnonpar(seq(0, 1, length = 10), data=bvdata,convex=T) #⼆元极值分布的依赖函数的⾮参数估计

abvnonpar(data = bvdata, method = "pick", plot = TRUE) #⼆元极值分布的依赖函数的⾮参数估计

abvevd(dep = M1["dep"], model = "log", plot = TRUE) #⼆元极值分布依赖函数的参数估计

abvnonpar(data = bvdata, add = TRUE, lty = 2) #⼆元极值分布的依赖函数的⾮参数估计

amvevd(dep = 0.5, model = "log") #多元极值分布依赖函数标准型

s3pts

amvevd(s3pts, dep = 0.5, model = "log") #多元极值分布依赖函数的参数估计amvevd(dep = 0.05, model = "log", plot = TRUE, blty = 1)

amvevd(dep = 0.95, model = "log", plot = TRUE, lower = 0.94)

asy

amvevd(s3pts, dep = 0.15, asy = asy, model = "alog")

amvevd(dep = 0.15, asy = asy, model = "al", plot = TRUE, lower = 0.7)

随机数

学会看病教学设计amvnonpar(s5pts, sdat, d = 5) #多元极值分布依赖函数的⾮参数估计amvnonpar(data = sdat, plot = TRUE)

amvnonpar(data = sdat, plot = TRUE, ord = c(2,3,1), lab = LETTERS[1:3]) amvevd(dep = 0.6, model = "log", plot = TRUE)

amvevd(dep = 0.6, model = "log", plot = TRUE, blty = 1)

bvdata

qcbvnonpar(c(0.9,0.95), data = bvdata, plot = TRUE)

qcbvnonpar(c(0.9,0.95), data = bvdata, epmar = TRUE, plot = TRUE) bvdata

hbvevd(dep = M1["dep"], model = "log", plot = TRUE)

uvdata

大气气溶胶trend

anova(M1, M2, M3) #⽅差分析

data(portpirie)

confint(m1) #参数的置信区间

pm1

plot(pm1) #

confint(pm1) #

uvdata

trend

anova(M1, M2, M3, M4)

plot(M2P)

rnd

fgev(uvdata, nsloc = data.frame(trend = trend, random = rnd))

fgev(uvdata, nsloc = data.frame(trend = trend, random = rnd), locrandom = 0)

uvdata

M1P

M2P

plot(M1P)

plot(M2P)

uvdata

大黄丹M1P

M1JP

plot(M1JP)

uvd

forder(uvd, list(mean = 0, sd = 1), distn = "norm", mlen = 365, j = 2)

forder(uvd, list(rate = 1), distn = "exp", mlen = 365, j = 2)

forder(uvd, list(scale = 1), shape = 1, distn = "gamma", mlen = 365, j =

forder(uvd, list(shape = 1, scale = 1), distn = "gamma", mlen = 365, j =

uvdata

anova(M1, M2)

par(mfrow = c(2,2))

plot(M1)

M1P

本文发布于:2024-09-21 03:34:45，感谢您对本站的认可！

本文链接：https://www.17tex.com/xueshu/4639.html

上一篇：活性炭碘吸附值的测定

下一篇：基于Pareto分布的风压极值计算方法

标签：分布函数极值

留言与评论（共有 0 条评论）