R help - Jul 2010 - Help needed with numericDeriv and optim functions

Hello All:


I have defined the following function (fitterma as a sum of exponentials)
that best fits my cumulative distribution. I am also attaching the
"xtime"
values that I have. I want to try two things as indicated below and am
experiencing problems. Any help will be greatly appreciated.


Best, Parmee

-----------------------


*fitterma <- function(xtime) { *

*a <- -0.09144115*

*b <- -0.01335756*

*c <- -2.368057*

*d <- -0.00600052*

*return(exp(a+b*xtime)+exp(c+d*xtime))*

*}*


I want to do two things:


*First, take the numerical derivative of this function (fitterma)* to obtain
probability density function. I tried using:


*> numericDeriv(fitterma,"xtime")*


*Error in numericDeriv(fitterma, "xtime") : *

*  cannot coerce type 'closure' to vector of type 'double'*


Looking at the complaint above, I don't understand how to use the above
function properly. Any help will be appreciated.


*Secondly, I want to minimize this function (fitterma) as a function of
xtime*. I tried using:


*> optim(1, fitterma)*


But I don't understand this function either of how to use it properly.


--------


*> xtime*

 [1]   1   2   3   4   7   8  10  11  12  13  15  16  18  19  20  21  22  24
 25  27  29  30  31  33  35  36  42

[28]  43  44  45  48  49  51  52  53  54  56  59  61  63  72  73  80  82  84
 87  90  92  95  99 100 103 105 110

[55] 111 112 117 118 122 126 132 133 139 140 143 144 151 153 156 162 164 177
186 200 201 216 228 231 242 250 260

[82] 278 283 287 314 340 357 378 384 389 392 411 467 553 587 991 999

	[[alternative HTML version deleted]]