Affiliation
American Association of Variable Star Observers (AAVSO)
Sun, 12/27/2015 - 05:38
Hi. I just did try to reproduce and recreate the procedure used in Vstar Manual under the Harmonics topic. I did load the most recent data of eta Aql (all) and performed a DC DFT standard scan on all Visual observations. When I obtained the top hits, I clicked the "Find Harmonics" button expecting to have 4 harmonics to be suggested by the software (This can be used as a decision support aid for model creation pg. 63, Vstar User Manual) and showing the same result from the manual as an example used but it's not. Instead, I am always getting the same result regardless of the object (have tried many variable stars) which always end up at 1 frequency harmonic. So my question is, is this normal or there's something wrong with my method?
File Upload
Hi Jonah
You said you loaded the latest (presumably last 2 years) of eta Aql data from AID.
The example in the user manual (page 59) suggests loading the JD range: 2455695.8431 to 2456419.8604
I just tried this with a DCDFT Standard Scan of the Visual band and I do see 4 harmonics in the Find Harmonics dialog.
Can you let me know what result you get when you load the above JD range?
I did notice that I get a slightly different period as the top-hit compared to what the manual says. I should look at that.
Something else I'm wondering about: what locale is your computer set to?
David
My bad. I loaded ALL the JD range instead of selecting the range in the manual but as what you've said, I still got different result. (P = 7.17874...)
Another question. Why is that I often get to have 1 harmonic when loading all the data in the set and using the find harmonics button instead of selecting certain JD range? I tried to apply only 1 harmonic and resulted in greater residuals in V band.
Thank you. This might clear my confusion.
And by the way, my system locale is set to United States
Your result agrees with mine. I think the example in that section of the manual are slightly out of sync with the text. I'll look at that.
We would need to look at some other specific examples to be able to say much more about the found harmonics. Let's talk about some specific examples where all data is loaded.
Note that you can of course add harmonics to the Fourier model irrespective of what Find Harmonics gives you, and as you say, this can improve residuals. Find Harmonics is a really just and aid and is not prescriptive. It really should at least provide a power value for each harmonic to be useful, since just because a harmonic was "searched", doesn't mean its presence was statistically significant in some way. This has been on my list for awhile: https://sourceforge.net/p/vstar/bugs-and-features/371/
David
I also meant to add that your screenshot shows you are using 2.16.8. The current version is 2.16.9 and 2.16.10 is in devevelopment/test and will be formally released in the near future.
David
Thank you for sharing your knowledge Sir David. It's been a while since I used Vstar again. I'll download the latest version later this time.
David
Thank you, sir!
David