我有代码
INJ.1<-"I01 I02 I03 I04 I05
2.78E+02 1.82E+03 3.62E+02 2.90E+02 7.73E+02
7.92E+02 1.21E+03 9.33E+02 6.32E+02 5.10E+02
2.30E+03 7.54E+02 9.60E+02 6.29E+02 1.05E+03
3.61E+03 3.05E+02 7.77E+02 5.87E+02 1.02E+03
3.89E+02 1.35E+03 7.66E+02 4.00E+02 7.43E+02
1.31E+03 1.63E+03 8.95E+02 3.85E+02 1.10E+02
1.39E+03 1.16E+03 9.07E+02 4.99E+02 2.48E+02
1.94E+03 1.09E+03 8.34E+02 5.22E+02 2.48E+02
2.04E+03 1.11E+03 7.85E+02 2.67E+02 4.27E+02
1.06E+03 1.36E+03 8.80E+02 6.13E+02 7.16E+02
1.40E+03 1.29E+03 8.65E+02 6.17E+02 9.79E+02
1.20E+03 1.68E+03 6.78E+02 6.10E+02 9.30E+02
1.45E+03 1.49E+03 7.66E+02 3.81E+02 1.07E+03
1.16E+03 1.58E+03 1.09E+03 5.33E+02 8.38E+02
1.33E+03 1.38E+03 9.10E+02 6.29E+02 8.80E+02
"
INJ<-as.matrix(read.table(text=INJ.1, header=T))
PRD.1<-"P01
981.32019
1062.5702
1439.7673
1694.0723
1085.1016
1243.6089
1191.5941
1302.2167
1333.5266
1242.0212
1342.6954
1371.2767
1394.1171
1400.7926
1373.1791
"
PRD<-as.matrix(read.table(text=PRD.1, header=T))
tao=as.matrix(c(1,1,1,1,1))
lambda=as.matrix(c(0.0251879,0.1599486,0.1812318,0.2626731,0.3355733,0.3221295,-1.3343501))
i.dash=matrix(ncol=ncol(INJ), nrow=(nrow(INJ)))
fn1 <- function (tao){
for (i in 1:ncol(INJ))
for (j in 1:nrow (INJ))
temp=0
for (k in 1:j)
i.dash[j,i]=(1/tao[i])*exp((k-j)/tao[i])*INJ[k,i]+i.dash[j,i]
target = abs(700-sum(colSums(i.dash)))
}
ini=c(1, 1, 1, 1, 1)
ans1<-optim(par=ini,fn1,hessian=TRUE)
我需要优化函数中显示的tao
的值。除了我注意到函数fn1
内的矩阵计算没有完成之外,代码保持给出相同的初始值。除了主要问题之外,我还有一个以上的问题是如何解决这个案例以实现o target
的最小值:
values
来准备目标函数,然后我对它们进行优化是否是R中可接受的方法?