[R-sig-ME] starting values are not correct length?

[Andrew Dolman]

Hi Michael,

You might be right and it's 3 parameters to estimate, or fewer.
Depends on which of those parameters are in the data frame and which
are variables to be estimated in the equation/model LVB.


andydolman at gmail.com



On 18 August 2010 17:42, Michael Larkin <mlarkin at rsmas.miami.edu> wrote:
> Can you help me understand this code?  How do you get 4 parameters instead of 3?  I see three parameters:  Lmax, t0, and K?  I guess Age is the fourth parameter?
>
> LVBbis.nlme=nlme(L~LVB(Age,t0,Lmax,K), data=datagr,
> fixed=list(t0~Pop-1,Lmax~Pop-1,K~Pop-1),
> random=t0+Lmax+K~1,
> start=list(fixed=c(-1.54,-1.54,710,700,0.21,0.21)))
>
>
>
> -----Original Message-----
> From: Andrew Dolman [mailto:andydolman at gmail.com]
> Sent: Wednesday, August 18, 2010 11:37 AM
> To: Michael Larkin
> Cc: r-sig-mixed-models at r-project.org
> Subject: Re: [R-sig-ME] starting values are not correct length?
>
> That error means that you've got the wrong number of starting values,
> not that the values themselves are wrong. Don't you have 4 parameters
> to estimate not 6?
>
>
>
> andydolman at gmail.com
>
>
>
> On 18 August 2010 17:28, Michael Larkin <mlarkin at rsmas.miami.edu> wrote:
>> I am trying to run a non-linear mixed effect model. I keep getting the
>> error of "starting values for the fixed component are not the correct
>> length" I tried different sized values for the starting values but keep
>> getting the same error message. Any advice would be greatly appreciated.
>> My code is below.
>>
>>
>>
>> LVBbis.nlme=nlme(L~LVB(Age,t0,Lmax,K), data=datagr,
>>
>> fixed=list(t0~Pop-1,Lmax~Pop-1,K~Pop-1),
>>
>> random=t0+Lmax+K~1,
>>
>> start=list(fixed=c(-1.54,-1.54,710,700,0.21,0.21)))
>>
>>
>>
>>
>>
>> Mike
>>
>>
>>    [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-mixed-models at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>>
>
>
>