问题描述
对于返回数据，我正在研究偏度和峰度对 cVaR 计算的重要性。我们首先比较一些分布，通过使用包“fitdistrplus”在 R 中使用 fitdist() 估计分布的参数。但是，我们希望针对不同数量的分布执行此操作（参见图片：SGT、GT、SGED、GED、t、norm）。
下面是 SGT 的示例代码，这里有一个问题：它为参数 p 和 q 的标准误差引入了 NaN。我也不知道如何准确选择起始值。
SGTstart <- list(mu=0,sigma=2,lambda = 0.5,p=2,q=8)
SGTfit_R <- fitdistrplus::fitdist(data = as.vector(coredata(R)),distr = "sgt",method = "mle",SGTstart)
summary(SGTfit_R)
数据样本以使其可重现：我的股票指数的返回向量
c("0","-1,008599424","0,73180187",443174024","-0,351935172",318784086","-2,171323799","1,270243431",761354019",417350946",906432976",066736422",867085373",119914361",300989601",482518259",787365385",443105439",318546686","-3,467674998",041540157",371281289",176752782",116893343",127522915",658070287",098348016",296391358",810635352",041779322",353974233",120090141",304927119",22772592",040768364",182218724",123136685",682709972",174093506",539704174",579080595",326346169",205503526",771928642",490828799",734822712",025733101",246531452",695585736",732413919",806417952",396105099",024558388",791232528",730410255",438890702",668400286",440996039",731823553",177515522",740085418",926248628",63516084",89996829",655117371",501033581",06526534",320866692",496350734",10157668",022333393",236934596",070586427",661662029",871334714",758891429",064748766",305132153",424033661",223444774",441840866",661390655",148399329",843067435",601099664",329590349",210791225",341341769",555892395",624026986",218851965",015859171",524283138",855634719",339281481",038507713",943784688",315857689",368982834",111684011",2409217",421815833",079319721",915338199",537387704",023004636",331854888",702733882",084343115",16901282",559404916",538587484",153683523",336562411",274946953",862901957",117407383",205205829",633347347",058712615",083562948",343190727",281380185",750972389",538678151",228222073",635385022",037379479",491444798",220272752",093162287",499512169",041394336",113330512",657485999",264647978",115056075",009763771",454629881",322398317",347112494",948127411",461194301",407013048",469481931",536045151",114726251",396772868",525885581")
最好的，恩乔
解决方法
答案是使用包 sgt