AS5001 (SUPAAAA) Advanced (Astronomical) Data Analysis

Keith Horne -- University of St.Andrews

10am Mon, Maths Theatre B

9am Tue & Thu, PandA Theatre C


Astronomy Picture of the Day .
Homework Set 1 (due Mon 3 Oct 2022)
Homework Set 2 (due Mon 24 Oct 2022)
Project 1 (due Wed 09 Nov 2022)
Project 2 (due Wed 23 Nov 2022)
PGPLOT (Graphics Subroutine Library): PGPLOT . Subroutine Descriptions . Quick Reference .
Example PGPLOT codes:
Source code: simple.for . Compile/link: simple.com . Resulting plots: simple.ps .pdf .gif
Source code: simplot.for . Compile/link: simplot.com . Resulting plots: simplot.ps .pdf .gif .
Numerical Recipes . Online . D.Hogg - Data Analysis Recipes: Probability Calculus for Inference .
Opinionated Lessons in Statistics by Bill Press.
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 , 41 , 47 , 48 , 49 .
The Ways of Our Errors (Incomplete Draft Textbook) woe.ps , .pdf .

2022 ADA Lectures & Videos
ada01, 6. 2022mp4. Astro datasets, noise, statistical/systematic errors, probability distributions, mean, median, variance, MAD, central moments, skewness, kurtosis, Parameterised PDFs (Uniform, Gaussian, Lorentzian, Poisson, Exponential, Chi^2).
ada02, 6. 2022mp4. 5-sigma rule, joint distributions, independence vs correlation, algebra of random variables, linear transformations, co-variance and correlation coefficients and matrices.
ada03, 6. 2022mp4. Spurious correlations, correlation vs causation, non-linear transformations, error bar propagation, bias correction, generating non-uniform random numbers, Box Muller transform.
ada04, 6, 2022mp4. Central Limit Theorem, convolution of Gaussians and Lorentzians, statistics, unbiased sample mean, weighted mean, inverse-variance weighted optimal average.
ada05, 6. 2022mp4. Optimal scaling, chi^2 fits, Delta_chi^2=1 criterion for 1-parameter uncertainty, distribution of chi^2_min and Delta chi^2, degrees of freedom.
ada06, 6, 2022mp4. Sample variance (unbiased estimator for sigma^2) and its uncertainty. Robust estimation, mean vs median, MAD, SANE, median filtering, sigma-clipping, various BoF statistics.
ada07, 6, 2022mp4. debiased estimator for powers of sigma, error bars are model parameters, maximum likelihood, parameters affecting sigma, fits to Poisson data. conditional probabilities, Bayes' theorem, Bayesian inference, gaussian priors.
ada08, 6. 2022mp4. Monte-Carlo and Bootstrap sampling, m-pameter k-sigma confidence regions, fit a line, correlated vs orthogonal parameters.
ada09, 6. 2022mp4. Linear Regression, Normal Equations, Hessian Matrix, Non-Linear Models, Linearisation, Amoeba (Downhill Simplex), MCMC (Markov-Chain Monte-Carlo).
ada10, 6. 2022mp4. Vector Space Perspective, Data Space Metric, Error bars in both X and Y.
ada11, 6. 2022mp4. Vector Space Methods. Scaling Orthogonal Patterns, Hessian Matrix for Non-Linear Models, Graham-Schmidt Orthogonalisation, Orthogonal Polynomials, Reduced Chi^2 for Rejecting Models, Occam's Razor, AIC, AICc, BIC.
ada12, 6. 2022mp4. Chi^2_min Diagnosis. Re-scaling error bars. Extra variance. Background fits - polys, splines, running optimal average, median filter, sigma clip, variance of median.
ada13, 6. 2022mp4. Timing analysis: ephemerides, O-C diagrams. Periodograms: spectral leakage, Fourier basis, Nyquist frequency, aliasing above Nyquist frequency, beating, pre-whitening, harmonics, data gaps.
ada14, 6. 2022mp4. Fourier analysis: amplitude and phase modulation, spike, white noise, dynamic power spectra. wavelets. spline decomposition. red vs white noise.
ada15, 6. 2021mp4. Cross-correlation analysis. Description of ADA Projects 1 & 2.
2021 ADA Lectures & Videos (to be replaced as we go along) -->

ADA Lectures (astro-specific) no longer given.
ada16 (handout only). High-Speed Photometry: photomultiplier tubes. coincidence losses. tagging. sky background fits. extinction and absolute calibration.
ada17 (handout only). CCD detectors. Bias and Flat Field calibration. CCD noise model: Readout noise, Gain. CCD Spectroscopy. Differential refraction. Tracing the spectrum. Sky subtraction. Normal vs Optimal extraction.
ada18 (handout only). Wavelength calibration. Extinction corrections including saturated Telluric lines. Absolute flux calibration. Slit loss corrections.
ada19 (handout only). Crowded Field CCD photometry.
ada20 (handout only). Tomography.
Return to Astronomy Group Homepage
Keith Horne (kdh1-nospam-@st-andrews.ac.uk)