We are excited to announce the launch of our new forums! You can access it forums.aavso.org. For questions, please see our blog post. The forums at aavso.org/forum have become read-only.
Its been about 3 months since I took the VStar class, and I don't remember the parameters. If I was faced with your dilema, I would go back and read the section on WWZ in the VStar Manual. If that does not answer your question, then perhaps someone who uses it regularly can help. You may have to post a little more info on what you are trying to do, and what your attempts have given.
The error you are seeing may well be a result of the gaps in the data. WWZ is brittle in the presence of gaps in the data. See ticket 328. It's not clear to me whether the error you are seeing is related to this, so I'll raise a separate ticket for this.
Thinking about what what you're trying to achieve, presumably determining whether there's period change over time, I looked at the variable type in the VSX Variable Types document:
SXPHE
Phenomenologically, these resemble HADS variables but they are pulsating sub-dwarfs of the spherical component, or old disk galactic population, with spectral types in the range A2-F5. They may show several simultaneous periods of oscillation, generally in the range 0.04-0.08 days, with variable- amplitude light changes that may reach 0.7 mag. in V. These stars are present in globular clusters.
I'm mulling this over. I wonder whether WWZ is the best tool here.
The phase plot at the VSX period produces a phase plot that is not very "tidy", I suppose because of the phrase I have highlighted above. I wonder whether creating phase plots for distinct "regions" of the naive phae plot using filters might be illuminating. I've done something similar for some RR Lyr stars.
I am indeed looking for period changes in CY Aqr. Over the past year, the data set shows many very good, continuous ~4hr time series with large time gaps or perhaps a few scattered data points inbetween. That seems to be the problem using WWZ. I was thinking about some sort of a windowed Fourier analysis plotting the Fourier periods from each good 4hr time series vs. time or perhaps O-C vs. time. Would you suggest plotting this in VStar using a simple file (substituting Period for Magnitude) or by some other method?
Yes, the section at the end of page 98 of the user manual applies here.
It's an interesting one Dave. For a short period, you may want to change the decay value for better frequency resolution. Then again, there will be a trade-off between this and temporal resolution.
I've mostly applied WWZ to LPVs.
Are you willing to post some data here so we can mull it over?
I have been experimenting with WWZ for CY Aqr which has a period of ~0.06 days. Because of the very short period, there are large data gaps, and the WWZ plot has large flat-bottomed regions. When I look at an area of a continuous time series (see attached ~ 4 cycles), then no matter what c parameters I enter (as recommended in the manual between 0.001 and 0.0125) I get the error code shown. Suggestions?
I don't know the answer to your question. As I understand it, one uses WWZ when the period is changing and one is looking to model that. These 4 cycle look pretty similar. I don't think that should matter. I believe there is an example in the help file/manual. Can you run the example on its data set, and get the result?
Hello Dave
Its been about 3 months since I took the VStar class, and I don't remember the parameters. If I was faced with your dilema, I would go back and read the section on WWZ in the VStar Manual. If that does not answer your question, then perhaps someone who uses it regularly can help. You may have to post a little more info on what you are trying to do, and what your attempts have given.
Gary
Hi Dave, Gary
The error you are seeing may well be a result of the gaps in the data. WWZ is brittle in the presence of gaps in the data. See ticket 328. It's not clear to me whether the error you are seeing is related to this, so I'll raise a separate ticket for this.
Thinking about what what you're trying to achieve, presumably determining whether there's period change over time, I looked at the variable type in the VSX Variable Types document:
I'm mulling this over. I wonder whether WWZ is the best tool here.
The phase plot at the VSX period produces a phase plot that is not very "tidy", I suppose because of the phrase I have highlighted above. I wonder whether creating phase plots for distinct "regions" of the naive phae plot using filters might be illuminating. I've done something similar for some RR Lyr stars.
David
Thanks David and Gary for your thoughts -
I am indeed looking for period changes in CY Aqr. Over the past year, the data set shows many very good, continuous ~4hr time series with large time gaps or perhaps a few scattered data points inbetween. That seems to be the problem using WWZ. I was thinking about some sort of a windowed Fourier analysis plotting the Fourier periods from each good 4hr time series vs. time or perhaps O-C vs. time. Would you suggest plotting this in VStar using a simple file (substituting Period for Magnitude) or by some other method?
- Dave Cowall
Hi Dave
I'll try to get O-C (https://sourceforge.net/p/vstar/bugs-and-features/235/) soon; that may be useful here.
Yes I too wondered about windowed Fourier analysis here.
David
Hi Dave, Gary
Yes, the section at the end of page 98 of the user manual applies here.
It's an interesting one Dave. For a short period, you may want to change the decay value for better frequency resolution. Then again, there will be a trade-off between this and temporal resolution.
I've mostly applied WWZ to LPVs.
Are you willing to post some data here so we can mull it over?
David
Thanks Gary and David -
I have been experimenting with WWZ for CY Aqr which has a period of ~0.06 days. Because of the very short period, there are large data gaps, and the WWZ plot has large flat-bottomed regions. When I look at an area of a continuous time series (see attached ~ 4 cycles), then no matter what c parameters I enter (as recommended in the manual between 0.001 and 0.0125) I get the error code shown. Suggestions?
- Dave Cowall
Hi David
I don't know the answer to your question. As I understand it, one uses WWZ when the period is changing and one is looking to model that. These 4 cycle look pretty similar. I don't think that should matter. I believe there is an example in the help file/manual. Can you run the example on its data set, and get the result?
Gary