Doctor en Ciencias, Mención Astronomía / Here I present an in-depth study of the central region of active galactic nuclei oriented to improve current mass estimation methods of distant super massive black holes and to infer the physical properties of the gas in their vicinity. In the first chapter I briefly introduce the basic concepts and present the most relevant discoveries and problematics associated to the central topic of this thesis. Then, in the second chapter, I present new calibrations of the so called single epoch black hole mass estimation method. This method uses emission lines from the broad line region such as the Halpha, Hbeta and MgII low ionization lines, and the CIV high ionization line. The novelty of this work is the usage of simultaneous observations of these emission lines that prevents from possible variability effects. The latter was possible thanks to the observations of 39 quasars a $z\sim1.55$ using the X-Shooter spectrograph of the VLT telescope whose wide spectral coverage allows simultaneous mapping of the aforementioned emission lines. In addition to presenting new calibrations, the results of this study indicate that low ionization lines provide more accurate mass estimations than CIV as it was suggested by previous studies. In the third chapter, I examine the possibility of improving current \CIV -based mass estimates of super-massive black holes by testing the performance of some methods proposed in the literature, including a method proposed in this thesis. All these methods are based on correlations found using small samples of less than 100 objects. In order to quantify the statistical robustness of these methods, in this work I use the Sloan Digital Sky Survey quasar database out of which I extracted a sample of nearly 30000 objects. The results suggest that the methods studied here have a very limited effect on the improvement of \CIV-based mass estimations. Finally, in the fourth chapter, I study the effect of gas distribution of the broad line region on mass estimations. This is possible thanks to the comparison between masses obtained from the single epoch method and those obtained from the fitting the accretion disc spectral energy distribution to standard accretion disk models. The results indicate a strong dependence of the ratio between both mass estimates with the observed width of the broad emission lines.