jcadien1 |
07/08/2015 05:37PM (Read 1586 times)
|
|
|
Status: offline
Registered: 11/28/2006
Posts: 21
|
I am trying to use the IRAF pdm period finder. I have begun by getting data on a known variable, with cataloged period of 86.5 days. Pdm seems to find a period, regardless of the min/max range I choose. The period is always between the min/max. So, if I set them to be 5-10 I get something around 7 for a period. If I set 80-90, I get 87. But there does not seem to be any figure to assess the prob that any of these is closer to correct than others. So, when using this with lightcurve data from a star with no known period, how can I decide?
Thanks, Jim
|
|
|
|
fitz |
07/08/2015 05:47PM
|
|
|
Status: offline
Registered: 09/30/2005
Posts: 4040
|
PDM will search for a period within the specified min/max window if you provide values, however it will determine it's own value if you leave these as INDEF. That's not to say this task will work well with your particular dataset, but if you don't specify the period window and the task comes up with something close to the know value then it might be appropriate. Unfortunately, PDM is the only period-finding task in the system, if it doesn't work well you may want to explore other tools.
|
|
|
|
jcadien1 |
07/08/2015 09:29PM
|
|
|
Status: offline
Registered: 11/28/2006
Posts: 21
|
If I set both minp and maxp to INDEF and run pdm I get the error that maxp <= minp. Does not seem to be affected by changing any other parameter, including auto selecting the range.
Jim
|
|
|
|
fitz |
07/08/2015 09:37PM
|
|
|
Status: offline
Registered: 09/30/2005
Posts: 4040
|
From the help page, it appears 0 (zero) is the magic default value for minp/maxp, not INDEF.
|
|
|
|
jcadien1 |
07/08/2015 09:48PM
|
|
|
Status: offline
Registered: 11/28/2006
Posts: 21
|
Sorry Fitz, but no luck. I had tried that earlier with no luck, but tried again, removing the INDEFs and substituting 0. No work, same error msg.
Jim
|
|
|
|
fitz |
07/08/2015 10:01PM
|
|
|
Status: offline
Registered: 09/30/2005
Posts: 4040
|
What about minp=0 and maxp=INDEF (or some equally large number) ?
|
|
|
|
jcadien1 |
07/09/2015 04:51PM
|
|
|
Status: offline
Registered: 11/28/2006
Posts: 21
|
Tried that and got a floating point invalid operation.
Jim
|
|
|
|
fitz |
07/09/2015 04:53PM
|
|
|
Status: offline
Registered: 09/30/2005
Posts: 4040
|
Task might not handle INDEF well, try maxp=1000000.
|
|
|
|
jcadien1 |
07/09/2015 09:52PM
|
|
|
Status: offline
Registered: 11/28/2006
Posts: 21
|
I tried that. pdm auto made minp=5000, and returned a period of 5135. Looks like it is not going to work.
Jim
|
|
|
|