[R] Trend significant at p=0.05?

[Bert Gunter]

This is not the place for a statistical discussion; it *is* the place
for R programming help.

However, you seem to be confused about confidence intervals and
significance, i.e. P values. You and others may wish to read about why
the "magic" P = .05 significance level is a **bad idea**, contributing
to bad science and irreproducible results, here:

http://www.nature.com/news/statisticians-issue-warning-over-misuse-of-p-values-1.19503?WT.mc_id=SFB_NNEWS_1508_RHBox

(The original ASA statement is linked, and other relevant references
are provided).

So in other words, you and all others who believe that P values have
scientific value may wish to reconsider that belief.

Best,

Bert
Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )


On Mon, Mar 14, 2016 at 8:49 AM, Chattopadhyay, Somsubhra
<sch298 at g.uky.edu> wrote:
> Hi,
>
> I am trying to calculate trends in my climatic time series, using the "Zyp"
> package. Initially, I tried with my daily data (I have daily rainfall data
> of 30 years) but then both the Zhang method and Yue-Pilon method yielded
> unrealistic results. The trend estimates were zero and so were the
> confidence limit bounds. But I guess, that is because the noise in the
> daily data. So, I converted my time series to annual data and now I see the
> attached output. I see that there is a positive trend, but my question is
> is this trend significant at 95% confidence interval? I suspect no because
> the confidence interval contains zero? Can anyone please confirm?
>
> I also attached the autocorrelation plot of the original time series.
>
> Thanks,
> Som
>
> --
> *Somsubhra Chattopadhyay*
> Graduate Research Assistant
> Biosystems and Agricultural Engineering Department
> University of Kentucky, Lexington, KY 40546
> Email: schattop14 at uky.edu
> Cell: 9198026951
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.