search for: lprobs

Displaying 3 results from an estimated 3 matches for "lprobs".

Did you mean: probs
2002 Feb 05
2
passing parameters in optim()
Hi all, I am using optim() to get estimate for parameter 'a' which will minimize the output of lprob.catch(a) and I got an error message: > optim(2,lprob.catch(x, freq,1, 0.5,0.5, a,- 0.7, 1, -0.7,0.1, 1,0.2,2,1,data.sf.mont)) Error in catch2.fun(k, f1, f2, a, b, A, B, R, n0, av, index.n) : Object "a" not found Why wasn't the initial value of a=2 passed to
2007 Oct 26
5
help
hello, please can anyone help me out. Am a new user of R program. Am having problem with this code below, not getting the expected results. 1. Each m, the cumulative sum should be 1.000 but the 2nd and 3rd m returned 2.000 and 3.000 instead of 1.000. 2. to get the LCL(m) and UCL(m) for each m base on these instructions if out.cum > 0.025 then LCL(m)= y-1 if out.cum >0.975
2007 May 11
0
EM covergence problem
...%Beta[(i-1)*2+1:2] v2 = Y%*%Beta[(i-1)*2+1:2] p1 = exp(v1)/(exp(v1)+exp(v2)) p2 = exp(v2)/(exp(v1)+exp(v2)) probs[,i] = ifelse (D==0,log(p1),log(p2)) } return (probs) } #H [individuals][class] E_step = function(alpha,Beta){#calc posterior of H tmpH = matrix(,nrow = 1000,ncol =3) lprobs = logProbInd(Beta) for(i in 1:3){#classes tmpH[,i] = alpha[i]*exp(lprobs[,i]) } H = tmpH /apply(tmpH,1,sum) return( H) } M_step = function(H,Beta){ #first part use direct estimation aita = apply(H,2,sum)/1000 opt.c = optim(Beta,Q2,H=H,method="BFGS",control = list(fnscale = -1)...