Hi
Attached is my data (2006_7) that im trying to run through PERIOD.
My procedure is as follows:
1) load/input the file "2006_7" (three columns of data, 1-time, 2-magnitudes, 3-mag errors) into slot 1,1
2) enter PERIOD/PERIOD
3) select frequency ranges: min=1, max=2, freq interval=1e-4
4) select slot 1,1 for input; 2,2 for output
5) run STRING (fills slot 2,2)
6) plot 2,2 as xwin (this plot is attached as "plot_1" (a jpeg file))
at this point all is fine and I then run PEAKS (slot 2,2; and whole freq range 0,0) and get a period for this data of:
Period = 0.6031726883406983 (days)
which is quite near what I expect.
Now, the next thing I am want to do is decrease the freq range, while keeping the freq interval the same:
1) freq: min=1.64, max=1.68, freq interval=1e-4
and then run STRING as per the same method as above. This then produces "plot_2" (attached).
In "plot_2", at the bottom left-hand corner there is an rogue datapoint that shouldnt be there. As this point has the shortest string length, it messes up the calculation of PEAKS.
My question thus revolves around why this rogue point is there, and why, when I do the same procedure as above, with a smaller freq range and different freq intervals (ie. 1e-5, 1e-3, etc) that point is still there.
Thanks to all!
Best,
Zach
----- Original Message ----
From: David Berry <[log in to unmask]>
To: [log in to unmask]
Sent: Wednesday, 22 April, 2009 16:33:15
Subject: Re: PERIOD
Hi Zach,
Some test data and an explicit script of commands that
demonstrates the problem would be really useful.
David
2009/4/22 zachariah cano <[log in to unmask]>:
> Hi
>
> I keep getting the same problem when I run PERIOD on a set of data I have obtained on an RR Lyrae star. Im trying to use string & peaks to determine a period and I keep encountering problems when I use different frequency ranges and different resolutions.
>
> For instance, I am expecting a period of about 0.603 days. When I use a freq range of 1-2 and a resolution of 1e-4 I get a decent plot, when when I change the freq interval to something smaller, say 1.64-1.68, and use any resolution (ie 1e-3, 1e-4, 1e-5, etc) when I run string, and then plot the result there is always a rogue datapoint at the bottom left-hand corner that then messes up the resulting peaks calculation.
>
> I have run the same data through IRAF task PDM and not had any problems in extracting a period, and I cannot understand whats going wrong with PERIOD.
>
> If I have not explained things clearly enough I also have plots of what I am describing.
>
> Thanks!
>
> Best,
> Zach
>
>
>
>
|