On 28 May 2015, at 10:17 , Martin Maechler wrote:
>
> On the other hand, to me, it did look a bit strange or "ugly" to always
> perform to different lm()s and the coefficients of one and the
> residuals of the other.
Yes, but the whole point was that if you do OLS on an integrated series, the
> Patrick Perry
> on Wed, 27 May 2015 23:19:09 -0400 writes:
{@PP, you forgot this part:}
> peter dalgaard
> on Thu, 21 May 2015 14:36:03 +0200 writes:
>> I suspect that what we really need is
>>
>> fitI <- lm(x ~ xreg - 1, na.action = na.omit)
>> f
> I suspect that what we really need is
>
> fitI <- lm(x ~ xreg - 1, na.action = na.omit)
> fit <- if(length(dx) > ncol(dxreg))
> lm(dx ~ dxreg - 1, na.action = na.omit)
> else list(rank = 0L)
> if(fit$rank == 0L) {
> ## Degenerate model. Proceed anyway so as not to break old code
> fit <- fit
Thanks for your help, Martin and Peter.
I tried taking the value of “n.used” from the C functions (ARIMA_CSS and
ARIMA_Like) instead of computing n.used on the R side. Here is a patch, in
case you’re interested:
https://github.com/patperry/r-source/commit/8fed79a6d2d558ef34738624a2a4f9e795bcf8
On 21 May 2015, at 12:49 , Martin Maechler
wrote:
>> peter dalgaard
>>on Thu, 21 May 2015 11:03:05 +0200 writes:
>
>> On 21 May 2015, at 10:35 , Martin Maechler
>> wrote:
>
I noticed that the 3.2.1 release cycle is about to start. Is there any
chance that this
> peter dalgaard
> on Thu, 21 May 2015 11:03:05 +0200 writes:
> On 21 May 2015, at 10:35 , Martin Maechler
wrote:
>>>
>>> I noticed that the 3.2.1 release cycle is about to start. Is there any
>>> chance that this fix will make it into the next version of R?
>
On 21 May 2015, at 10:35 , Martin Maechler
wrote:
>>
>> I noticed that the 3.2.1 release cycle is about to start. Is there any
>> chance that this fix will make it into the next version of R?
>>
>> This bug is fairly serious: getting the wrong variance estimate leads to
>> the wrong log-like
> I noticed that the 3.2.1 release cycle is about to start. Is there any
> chance that this fix will make it into the next version of R?
>
> This bug is fairly serious: getting the wrong variance estimate leads to
> the wrong log-likelihood and the wrong AIC, BIC etc, which can and does
> lead t
I noticed that the 3.2.1 release cycle is about to start. Is there any
chance that this fix will make it into the next version of R?
This bug is fairly serious: getting the wrong variance estimate leads to
the wrong log-likelihood and the wrong AIC, BIC etc, which can and does
lead to suboptimal
The bug repository is like an elephant: It doesn't forget, but the gestation
period is long.
In the present case, it is clear that something is not right, but someone needs
to have sufficient recall and insight to check that your proposed fix is not
unfixing a deliberate change. We should get t
10 matches
Mail list logo
