Observational constraints on the structure and evolution of quasars

I use X-ray and optical data to investigate the structure of quasars, and its dependence on luminosity, redshift, black hole mass, and Eddington ratio. In order to facilitate my work, I develop new statistical methods of accounting for measurement error, non-detections, and survey selection functions. The main results of this thesis follow. (1) The statistical uncertainty in the broad line mass estimates can lead to significant artificial broadening of the observed distribution of black hole mass. (2) The z = 0.2 broad line quasar black hole mass function falls off approximately as a power law with slope ∼ 2 for MBH ≳ 108 M⊙ . (3) Radio-quiet quasars become more X-ray quiet as their optical/UV luminosity, black hole mass, or Eddington ratio increase, and more X-ray loud at higher redshift. These correlations imply that quasars emit a larger fraction of their bolometric luminosity through the accretion disk component, as compared to the corona component, as black hole mass and Eddington ratio increase. (4) The X-ray spectral slopes of radio-quiet quasars display a non-monotonic trend with Eddington ratio, where the X-ray continuum softens with increasing Eddington ratio until L/LEdd ∼ 0.3, and then begins to harden. This observed non-monotonic trend may be caused by a change in the structure of the disk/corona system at L/LEdd ∼ 0.3, possibly due to increased radiation pressure. (5) The characteristic time scales of quasar optical flux variations increase with increasing MBH, and are consistent with disk orbital or thermal time scales. In addition the amplitude of short time scale variability decreases with increasing MBH. I interpret quasar optical light curves as being driven by thermal fluctuations, which in turn are driven by some other underlying stochastic process with characteristic time scale long compared to the disk thermal time scale. The stochastic model I use is able to explain both short and long time scale optical fluctuations.