[R-SIG-Finance] rugarch: Mean model used for armaOrder(0,0)

[alexios galanos]

In the limit they should be.

You may want to read this section:

https://bookdown.org/compfinezbook/introcompfinr/maximum-likelihood-estimation.html


If you want to impose that the mean equals the sample mean, you can set the parameter as fixed.

Alexios



> On May 4, 2023, at 1:57 PM, Simon Rhodes <si.g.rhodes using gmail.com> wrote:
> 
> Is there a reason why the estimate for the mean would be different from the sample mean of the data? I seem to be getting results for mu, when using armaOrder(0,0) and include.mean=true, which are relatively different from the sample mean of my data. Similarly, when I simulate that model I get a simulation mean close to mu and not the sample mean of the data.
> 
> Thanks,
> Simon
> 
> On Tue, May 2, 2023 at 3:54 PM Alexios Galanos <alexios using 4dscape.com <mailto:alexios using 4dscape.com>> wrote:
>> Yes, the additional argument ‘include.mean’ which defaults to TRUE does exactly that, estimating this value.
>> 
>> If you are estimating a GARCH model with no ARMA terms then you may also like to try out the newly released tsgarch package (https://github.com/tsmodels/tsgarch)
>> 
>> Alexios
>> 
>> 
>>> On May 2, 2023, at 11:25, Simon Rhodes <si.g.rhodes using gmail.com <mailto:si.g.rhodes using gmail.com>> wrote:
>>> 
>>> I had a question about what model is used for the mean when armaOrder is
>>> set to (0,0). Am I correct in assuming it is simply the mean of the data?
>>> If not, is there a way to specify that the mean of the sample should be
>>> used as the mean for the model?
>>> 
>>> 
>>> Thanks,
>>> Simon
>>> 
>>>    [[alternative HTML version deleted]]
>>> 
>>> _______________________________________________
>>> R-SIG-Finance using r-project.org <mailto:R-SIG-Finance using r-project.org> mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
>>> -- Subscriber-posting only. If you want to post, subscribe first.
>>> -- Also note that this is not the r-help list where general R questions should go.


	[[alternative HTML version deleted]]