<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
By using fmle you are trying to fit a model with 0 parameters.<br>
<br>
The best way to calculate the likelihood conditional on a & b is
to use the logl function slot in FLSR, e.g.<br>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<span class="Apple-style-span" style="border-collapse: separate;
color: rgb(0, 0, 0); font-family: Arial; font-style: normal;
font-variant: normal; font-weight: normal; letter-spacing: normal;
line-height: normal; orphans: 2; text-align: auto; text-indent:
0px; text-transform: none; white-space: normal; widows: 2;
word-spacing: 0px; -webkit-border-horizontal-spacing: 0px;
-webkit-border-vertical-spacing: 0px;
-webkit-text-decorations-in-effect: none;
-webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px;
font-size: medium; "><span class="Apple-style-span"
style="font-family: Monospace; font-size: 13px; line-height:
17px; text-align: -webkit-left; white-space: pre-wrap; ">
<pre tabindex="0" class="GD40030CKR" style="font-family: Monospace; font-size: 10pt !important; outline-style: none; outline-width: initial; outline-color: initial; border-top-style: none; border-right-style: none; border-bottom-style: none; border-left-style: none; border-width: initial; border-color: initial; white-space: pre-wrap !important; margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; line-height: 1.3; ">
data(ple4)
pSR=fmle(as.FLSR(ple4,model="bevholt"))
<span class="GD40030CBR ace_keyword" style="color: blue; ">pSR@logl(params(pSR)["a"],params(pSR)["b"],ssb(pSR),rec(pSR))</span>
this is different from the logLik slot that just contains the value that fmle returned.
I did suggest that we tidy this up, e.g.
computeLogLik() would be a friendly wrapper around <span class="GD40030CBR ace_keyword" style="color: blue; ">pSR@logl</span><span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Arial; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; font-size: medium; "><span class="Apple-style-span" style="font-family: Monospace; font-size: 13px; line-height: 17px; text-align: -webkit-left; white-space: pre-wrap; "></span></span>
Not sure if it was ever implemented, but I can easily add it.
Laurie
</pre>
</span></span><br>
On 02/16/2012 11:09 AM, Hintzen, Niels wrote:
<blockquote
cite="mid:1356C6E49EEAF34A8745C1E0293643A42B574C0739@scomp0536.wurnet.nl"
type="cite">
<pre wrap="">Hi,
Trying to calculate the likelihood of a SR fit using fmle, but this fails..
The a and b parameters are obtained from a converged fit. Any idea what is going on here and how I can fix it?
Thanks a lot. Using FLCore 2.4 on R2.13.2 in 32-bit mode.
Niels
fmle(SR.function,fixed=list(a=4.6,b=1.56))@details$value
Nelder-Mead direct search function minimizer
function value for initial parameters = -12.689450
Scaled convergence tolerance is 1.89088e-07
Stepsize computed as 0.100000
Exiting from Nelder Mead minimizer
1 function evaluations used
Error in do.call("[<-", c(list(x = <a class="moz-txt-link-abbreviated" href="mailto:x@.Data">x@.Data</a>), dx, list(value = value))) :
replacement has length zero
</pre>
</blockquote>
<br>
</body>
</html>