You replied only to me. But I'm replying to you directly and the r-help
list as well. Hopefully someone there will be able to help you with this
statistical methodology. Please remember that I'm very ignorant of
statistics (I made a C in it in college). Your reply did seem quite
understandable to me, given my abysmal ignorance of statistics. Given that,
I am bowing out of this discussion because I would only add confusion and
not enlightenment.
On Sat, Aug 23, 2014 at 10:25 PM, 西风古道 <lx900...@qq.com> wrote:
> I would like to know hou to do the recursive regression in R. Suppose
> there are 400 observations. The regression starts at the r1th of the
> sequence and ends at r2th. First let the regression sequence start at 1st
> obesrvations and end at 40th. We get the first t-statistics from that
> regression. Then let the end point increase 1 in turn. The starting point
> also varies. For instance, during the second regression , first we run the
> regression via the sample 1-41 and then we run the regression via the
> sample 2-41. We get two t-statistics and we choose the max of the two. If
> the end point is 42, we choose the maximum t-statistics from the following
> three regression : sample 1-42,2-42,3-42.
>
>
> ------------------ 原始邮件 ------------------
> *发件人:* "John McKown";<john.archie.mck...@gmail.com>;
> *发送时间:* 2014年8月24日(星期天) 上午9:18
> *收件人:* "西风古道"<lx900...@qq.com>;
> *抄送:* "r-help"<r-help@r-project.org>;
> *主题:* Re: [R] How can I let the dimension change via the circulation?
>
> Apparently part of your message was removed by the email software. You
> MUST NOT use HTML encoded messages on this forum because the list software
> removes all HTML encoded information before sending it to us. It appears
> that the Matlab code was a victim of this removal. Also, don't try to paste
> a "screen snap shot" or "picture". You really need to paste in just plain
> text.
>
> I tried to make a decent guess, but the only thing I can find that _might_
> be what you are asking about is the Agumented Dickey-Fuller Test. I don't
> even know what that _is_, even after reading the Wikipedia article. But
> that article mentioned the "tseries" package in R. It contains a function
> called "adf.test()". I don't know if this is what you need or not. But you
> can review it by installing "tseries" by entering the R command:
> install.packages("tseries") at the R prompt. I am a system admin &
> programmer, not a statistician or economist. What I read:
> http://en.wikipedia.org/wiki/Augmented_Dickey%E2%80%93Fuller_test . But I
> don't understand it.
>
>
> On Sat, Aug 23, 2014 at 7:22 AM, 西风古道 <lx900...@qq.com> wrote:
>
>> Here is the fraction of Matlab code for the Generalized Sup ADF Test for
>> the bubble testing:
>>
>> The dimension of the matrix rwadft may vary via the circulation. I then
>> imitate this code by the R program as the following:
>> for (dim0 in 1:30){
>> rwadf<-numeric(dim0)
>> ....
>> in the circulation but it does not work. The dimension of the rwadf do
>> not vary and I can not get the right answer.
>> I want to know if I can let the dimension of the rwadf change via the
>> circulation. How should I write the R code?
>> [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-help@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>
>
>
>
> --
> There is nothing more pleasant than traveling and meeting new people!
> Genghis Khan
>
> Maranatha! <><
> John McKown
>
--
There is nothing more pleasant than traveling and meeting new people!
Genghis Khan
Maranatha! <><
John McKown
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
