Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Lowess line based on different time frame to graph x-axis


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Lowess line based on different time frame to graph x-axis
Date   Tue, 11 Sep 2012 10:21:50 +0100

This is the way that -lowess- works. Despite weighting, the essence of
the method is working with data points within a window, so the
"smooth"  can easily show jumps as individual data points are, or are
not, within the window.

I don't understand what kind of fudge you are seeking, but I suggest
that it's best to choose another smoother. In this particular case,
it's arguable that a smoother that doesn't treat past and future
symmetrically is preferable. -tssmooth- has various possibilities.

Regressing values on previous values and time often is a very smooth smoother.

Nick

On Tue, Sep 11, 2012 at 10:07 AM, Mathew Wilkins <MWilkins@unfccc.int> wrote:

> I have a series of monthly reports where we are trying to show a trend, the
> lowess line is based upon a monthly figure.  As a result the line is not
> completely smooth (the line is 'jumping' around e.g. 2012m1 = 134, 2012m2 =
> 158) I am looking for a way to remove the ‘jumps’ from the lowess lines.
>
> Is there a way where a lowess line in a graph can be based on a weekly time
> frame but keep the x-axis on the graph showing in monthly time frame?

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index