Convert optimization problem from R to Matlab

Question

Carlos Bosley 2019-7-25

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/473507-convert-optimization-problem-from-r-to-matlab

评论： Carlos Bosley 2019-7-25

I have the following code from R:

fleishtarget<-function(x,a){
  ##
  #The target function for solving equations 18, 19, and 20#
  # from page 523 of Fleishman.#
  #It does this by changing the system of three equations into a #
  # minimization problem.  Set the equations equal to zero, square,#
  # and sum up.  The b, c, and d that minimize the set of equations # 
  # at zero must also solve the three individually.#
  ##
  b<-x[1]
  cc<-x[2]
  d<-x[3]
  g1<-a[1]
  g2<-a[2]
  (2 - ( 2*b^2 + 12*b*d + g1^2/(b^2+24*b*d+105*d^2+2)^2 + 30*d^2 ) )^2 +
    (g2 - ( 24*(b*d+cc^2*(1+b^2+28*b*d)+d^2*(12+48*b*d+141*cc^2+225*d^2)) ) )^2+
    (cc - (g1/(2*(b^2+24*b*d+105*d^2+2)) ) )^2
}
findbcd<-function(skew,kurtosis){
  ##
  #Uses the built in minimization function to solve for b, c, and d#
  # if the skew and kurtosis are given. Try findbcd(1.75,3.75) and#
  # compare to Table 1 on page 524 of Fleishman.
  ##
  optim(c(1,0,0),fleishtarget,a=c(skew,kurtosis),method="BFGS",
        control=list(ndeps=rep(1e-10,3),reltol=1e-10,maxit=1e8))
}