American Association of Variable Star Observers (AAVSO)
Sun, 09/30/2018 - 02:54

I’m trying to look at phase modulation using kepler data. I’ve read several papers concerning this and it looks pretty straight foreward. The problem I’m having is that if I restrict the frequency range, I get very different values for the phase. The frequencies and amplitudes change very little.
To make this more clear, the star I’m looking at apparently has both g and p modes. I only want to see the modulation on the p modes (g modes aren’t very good for this type of analysis). There are three dominant p modes ( maybe rotationally split) and two dense sets of smaller frequencies ( g modes(?) and beat frequencies), some of which are comparable to the dominant modes.
For example, using period04 I get phase values for the phase of: 0.989782662, if I use frequency range of 0-nyquist, 0.718183, if I use frequency range 0.02-nyquist, and 0.088020852 if I use frequency range 10-nyquist.
What is wrong? When you add waves together, you add pointwise and it is commutative.
This makes me suspicious of some of my other calculations and of what I’ve read. I often see frequencies restrictions quoted as very low are often instrumental or rotational so are excluded.

Any suggestions will be appreciated.