[Aavso-photometry] Software for combining time series data takenwith different filters to calculate transformed magnitudes

[Michael Newberry]

Lionel, Your situation is not that unusual. At some point many people 
involved in research will reach a roadblock where no "canned" software will 
do exactly what they need. In that situation, it is good to learn a high 
level computing language or a scripting language that leverages as many 
existing (and relevant) software tools as possible. I thnk you are at that 
point. For example, you mentioned wanting to take into account images that 
had a large PSF because the seeing flared, or those having high errors, or 
whatever. That adds complexity, but it certainly is possible to do.

Michael

----- Original Message ----- 
From: "Lionel Catalan" <lcatalan at lakeheadu.ca>
To: "'Michael Newberry'" <mnewberry at mirametrics.com>
Sent: Friday, September 05, 2008 1:45 PM
Subject: RE: [Aavso-photometry] Software for combining time series data 
takenwith different filters to calculate transformed magnitudes


> Of course, you are right that it would help me to learn a bit of
> programming. The issue has always been the availability of time to do so.
> But in this case there may not be any other way around it...
>
>
> Lionel
>
> -----Original Message-----
> From: Michael Newberry [mailto:mnewberry at mirametrics.com]
> Sent: Friday, September 05, 2008 4:38 PM
> To: Lionel Catalan; aavso-photometry at mira.aavso.org
> Subject: Re: [Aavso-photometry] Software for combining time series data
> takenwith different filters to calculate transformed magnitudes
>
> There are various ways to do what you want and some parts of it may be
> specific to the kind of project you are doing. I'm not sure I would mess
> with Excel, but that's my way of doing things. In my opinion, the best 
> path
> would be to write a simple script, "customized" to your particular 
> project's
> needs, that pulls the reduced photometry from a table (or tables) and
> interpolates it to a new time basis. In the end, I think a little light
> programming would be worthwhile to learn.
>
> Michael Newberry
>
> ----- Original Message -----
> From: "Lionel Catalan" <lcatalan at lakeheadu.ca>
> To: "'Michael Newberry'" <mnewberry at mirametrics.com>
> Sent: Friday, September 05, 2008 1:17 PM
> Subject: RE: [Aavso-photometry] Software for combining time series data
> takenwith different filters to calculate transformed magnitudes
>
>
>> Hi Michael,
>>
>> I'm studying high mass X-ray binaries, and the time scale of changes
>> is greater than a few hours, whereas my images are taken every minute
>> or less.
>> I would think that I could start by linear interpolation, since it
>> would be the simplest. My problem is that I am not a programmer, and I
>> don't see how I can easily implement an effective interpolation scheme
>> in excel. I have hundreds of data points every night, and I obviously
>> cannot manually interpolate between each couple of successive
>> observations for each filter.
>>
>> Lionel
>>
>>
>> -----Original Message-----
>> From: Michael Newberry [mailto:mnewberry at mirametrics.com]
>> Sent: Friday, September 05, 2008 4:05 PM
>> To: Lionel Catalan; aavso-photometry at mira.aavso.org
>> Subject: Re: [Aavso-photometry] Software for combining time series
>> data takenwith different filters to calculate transformed magnitudes
>>
>> Be careful about interpolation according to the kind of science you
>> are trying to do. There's no patent answer for this.
>>
>> First of all, a general rule would be to interpolate only if the
>> timescale of what you are studying does not vary significantly on the
>> timescale of your observations. You want the data points to conincide
>> with what is happening on the star. For example, if doing RR Lyr, UV
>> Ceti flare stars asteroid rotation, etc., then I would use the actual
>> observation times and not interpolated ones (but see item 2, below).
>> OTOH, if you are doing Mira variables, etc., having a very slow rate
>> of change, then I don't think anyone would care about interpolating near
> the time of observation.
>>
>> Second, what kind of interpolation would you use? Would you thread the
>> points, do a piece-wide linear interpolation, cubic spline, etc.? Each
>> will give you a different magnitude value at whatever interpolated
>> time you choose. Be sure you understand the effects of the method you
>> use so that you get numbers that are scientifically defendable.
>>
>> Michael Newberry
>>
>> ----- Original Message -----
>> From: "Lionel Catalan" <lcatalan at lakeheadu.ca>
>> To: <aavso-photometry at mira.aavso.org>
>> Sent: Friday, September 05, 2008 12:49 PM
>> Subject: [Aavso-photometry] Software for combining time series data
>> takenwith different filters to calculate transformed magnitudes
>>
>>
>>>I have times series data for B,V, and I filters that I would like to
>>>combine  to calculate transformed magnitudes using transformation
>>>equations. I use  a  filter wheel to change filters between images, so
>>>I have sequences of  images  such as bvibvibvibvi etc.  Some images
>>>have to be rejected from time to  time  due to wind or clouds, so the
>>>sequences are not exactly regular. I already  have determined the
>>>transformation coefficients for my system.
>>> Essentially,
>>> I think that I need a program (or perhaps an excel macro) that can
>>>interpolate instrumental magnitudes for each filter to common times so
>>>that  consistent values of b-v and v-i can be calculated.  I would
>>>appreciate  any  suggestion.
>>>
>>> Lionel CTE
>>> _______________________________________________
>>>
>>> Aavso-photometry mailing list
>>> Aavso-photometry at mira.aavso.org
>>> To change options or unsubscribe, goto
>>> http://www.aavso.org/mailman/listinfo/aavso-photometry
>>>
>>>
>>
>>
>>
>
>
>