Re: st: possible fourier transformation

[Austin Nichols <austinnichols@gmail.com>]

Fitzmaurice, Ann E. <a.e.fitzmaurice@abdn.ac.uk> :

There is almost certainly a problem with your data; if those are
adjacent years, there is no way the rate of prenatal care drops from
78 to 20, rises to 99, then drops back to 40 in the space of a few
years.  That said, the transform you want, constructing 1+floor(T/2)
series of different frequencies from one series of length T, is
described in detail on page 334 in

Baker, Michael, Dwayne Benjamin, and Shuchita Stanger. 1999.
"The Highs and Lows of the Minimum Wage Effect: A Time-Series
Cross-Section Study of the Canadian Law."
Journal of Labor Economics, 17(2): 318-350.
http://www.jstor.org/stable/2660649

and you will also find this helpful:
http://www.stata-journal.com/sjpdf.html?articlenum=st0116

On Tue, Apr 20, 2010 at 4:26 PM, Fitzmaurice, Ann E.
<a.e.fitzmaurice@abdn.ac.uk> wrote:

> I have a series of data points that  represent different attributes along the continuum of pregnancy related services, so for example I have the percentage of women who had antenatal care, the percentage of women who delivered in an institution etc
>
> I have this data for a series of countries and there are at present 12 attributes which describe this continuum
>
> I have also within each country the same series of attributes by the wealth status of the family (five series)
>
> What I would like to do is to determine if the distributions, either between countries , or within countries by the wealth status are different (significantly )
>
> I would also like to generate some summary statistic that describes the distribution
>
> It has been suggested that I look at a fourier transformation , I know that stata can do this, but
>
> a) This is a new process to me and therefore do not know how to carry this out in stata
> b) Once the transformation is carried out ,  what would the next steps in the analysis
>
> The series for one of the countries is :
> 73.8
> 78.1
> 63.2
> 29.5
> 20.6
> 19.4
> 33.0
> 98.9
> 93.3
> 82.4
> 36.2
> 40
> All   of the above figures are percentages
>

*
*   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/