data analysis do it yourself!. what to do with your data? report it to professionals (e.g., aavso)...

37
Data Analysis Do it yourself!

Upload: jennifer-thomson

Post on 27-Mar-2015

213 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Data Analysis

Do it yourself!

Page 2: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

What to do with your data?

• Report it to professionals (e.g., AAVSO)– Excellent! A real service to science; don’t

neglect this

• Publish observations (e.g., JAAVSO)

• Analyze it – yourself!

Page 3: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

But …

• I’m not a mathematician– Let the computer do the math

• I’m not a programmer– Get programs from the net (often free)

• I don’t know how to use or interpret them– Neither do the pros!– Practice, practice, practice …

Page 4: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Time Series Analysis

• A time series is a set of data pairs

• t is the time, x is the data value• Usually, times are assumed error-free• Data = Signal + Error

• x can be anthing, e.g. brightness of variables star, time of eclipse, eggs/day from a laying hen

Naxt aa ,...,3,2,1),,(

)()( tftx

Page 5: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Basic properties of data x

Actual

• Mean = = expected value

• Standard deviation = expected rms difference from mean

Estimated

• Average = estimated

• Sample standard deviation = estimated

Page 6: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Average and sample standard deviation

• Average

• Sample standard deviation

xN

x1

2)(1

1xx

Ns

Page 7: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Method #1: world’s best• Eye + Brain: Look at the data!

• Plot x as a function of t: Explore!

• Scientific name:

Visual Inspection

• World’s best – but not infallible

• Programs:– TS http://www.aavso.org– MAGPLOT http://www.aavso.org

Page 8: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Method #2: Fourier Analysis

• Period analysis and curve-fitting

• Powerful, well-understood, popular

• Programs– TS http://www.aavso.org– PerAnSo http://www.peranso.com

Page 9: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Method #3: Wavelet Analysis

• Time-frequency analysis

• Old versions bad, new version good

• Programs:– WWZ http://www.aavso.org– WinWWZ http://www.aavso.org

Page 10: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Visual Inspection

Page 11: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Let’s take a look

Page 12: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Fourier Analysis

Page 13: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Fourier analysis for period search

• Match the data to sine/cosine waves

• = frequency

• Period =

• Amplitude = A = size of fluctuation

• Obvious choice is period; mathematically sound choice is frequency

)2sin()2cos()( 210 tctcctf

/1P

Page 14: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Null Hypothesis (important!)

• Null hypothesis: no time variation at all

• So = constant

• So,

• Quite important! Often neglected. Even the pros often forget this.

)(tf

aax

Page 15: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Is it real?

• Fit produces a test statistic under the null hypothesis

• Is usually “ /degree of freedom” (d)

• Linear: is significant (not just by accident) at 95% confidence

• 95% confidence means 5% false-alarm probability

2

42

Page 16: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Meaning of significance

• Significance does not mean the signal is linear, sinusoidal, periodic, etc.

• It only means the null hypothesis is incorrect, i.e., the signal is not constant

• Important!!!

Page 17: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Pre-whitening

• If you find a significant fit, then subtract the estimated signal, leaving residuals

• Analyze the residuals for more structure

• This process is called pre-whitening

Page 18: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

How to choose frequency?

• Test all reasonable values, get a “strength of fit” for each. Common is “chi-square per degree of freedom” (but there are many)

• Plot frequency .vs. fit – the Fourier transform (aka periodogram, aka power spectrum)

Page 19: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Fourier decomposition

Any periodic function of period P

(frequency ) can be expressed as a Fourier series:

P/1

...)2cos()2sin(

...)6cos()6sin(

)4cos()4sin(

)2cos()2sin()(

33

22

11

tnctnb

tctb

tctb

tctbatF

nn

Page 20: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Fundamental + harmonics

For a pure sinusoid, expect response at frequency

For a general periodic signal at a given frequency, expect a fundamental component at , as well as harmonics at frequencies etc.

,4,3,2

Page 21: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Lots of Fourier methods

• FFT: fast Fourier transform– Not just fast: it’s wicked fast

– Requires even time spacing

– Requires N=integer power of 2

– Beware!

• DFT: discrete Fourier transform– Applies to any time sampling, but incorrect results for

highly uneven (as in astronomy!)

– Beware!

Page 22: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Problems from uneventime sampling

• Aliasing: false peaks, often from a periodic data density

• Aliases at

• Common in astronomy: data density have a period P = 1 yr = 365.2422 d, so

• Solution: pre-whitening

datasignal n

002738.0data

Page 23: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Aliasing

Page 24: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Aliasing: UZ Hya

Page 25: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Problems from uneventime sampling

• Mis-calculation of frequency (slightly) and amplitude (greatly); sabotages prewhitening

Page 26: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Solution: better Fourier methods (for astronomy)

• Lomb-Scargle modified periodogram– Improvement over FFT, DFT

• CLEAN spectrum– Bigger improvement

• DCDFT: date-compensated discrete Fourier transform (this is the one you want)

• CLEANEST spectrum: DCDFT-like for multiple frequencies

Page 27: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

DCDFT

• Much better estimates of period, amplitude

Page 28: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Let’s take a look

• Peranso (uses DCDFT and CLEANEST)

• Available from CBA Belgium– http://www.peranso.com

Page 29: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect
Page 30: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect
Page 31: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Fourier transform (CLEANEST) of TU Cas

Page 32: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Wavelet Analysis

Page 33: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Wavelets

• Fit sine/cosine-like functions of brief duration

• Shift them through time

• Gives a time-frequency analysis

Page 34: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Problems

• Same old same old: uneven time spacing, especially variable data density, invalidate the results

• But: even worse than Fourier

• Essentially useless for most astronomical data

Page 35: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Wavelet methods

• DWT: discrete wavelet transform– Just not right for unevenly sampled data

(astronomy!)

• Solution: WWZ = weighted wavelet Z-transform

Page 36: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Let’s take a look

Page 37: Data Analysis Do it yourself!. What to do with your data? Report it to professionals (e.g., AAVSO) –Excellent! A real service to science; dont neglect

Data Analysis• Do it yourself

• Use your eyes and brain

• Healthy skepticism

[email protected]

• Enjoy!