Fitting

• For numerical way try doubling the x range
• look at running the fit for more iterations to get correct value
• wings have some other complicated structure
• restrict and to be positive and put bounds on parameters
• run the fit to get good guesses to speed up the fitting process
• Look at the published vega lines and determine whether parameters are similar
• look at eloide [http://atlas.obs-hp.fr/elodie/] spectra,
  • look at data file
  • convolute with our instrumental response and compare to our fit
• Plot a vertical dotted line for the real value of
• Look at how scales with the central value of the line
• look at finding physical parameters from the and

Errors

• think about what is causing the fluctuation within the lamp
• Look at using a larger gain, as could be an underestimate
• could be the noise of the electronics,
  • Look at the errors when the CCD is cooled right down

-- JosephBayley - 29 Jan 2016