The presence of the warm absorber of considerable optical depth is seen in many AGN. We show that this medium may affect significantly the optical/UV spectrum of an AGN by backscattering a fraction of the total radiation flux towards the disc surface. We consider in detail the case when the disk extends down to a marginally stable orbit, all the emission comes from the disk surface and the scattering medium forms a cone around the symmetry axis. Disc irradiation results in much flatter optical/UV continuum than predicted by standard disc models. The effect depends both on the total optical depth of the warm absorber and on the specific density distribution of this medium so the analysis of the optical/UV continuum allows to obtain constraints for the warm absorber complementary to those obtained from the soft X-ray data analysis. We give results for two exemplary sources - RE J1034+396 and PG1211+143, and for the bluest composite quasar spectrum of Richards et al. obtained from SDSS.

We construct a quasar extinction curve based on the blue and red composite quasar spectra of Richards et al. (2003) prepared from the SDSS survey. This extinction curve does not show any traces of the 2200 A feature characteristic of the Interstellar Medium, and this indicates that graphite grains are likely absent close to quasar nuclei. The extinction is best modeled by AC amorphous carbon grains, assuming a standard distribution of grain sizes (p=3.5) but slightly larger minimum grain size (a_{min} = 0.016 microns) and lower maximum grain size (a_{min} = 0.12 microns) than the respective canonical values for the interstellar medium. The dust composition is thus similar to that of the dust in AGB stars. Since graphite grains form from amorphous carbon exposed to strong UV irradiation the results indicate that either the dust forms surprisingly far from the active nucleus or in a wind that leaves the nucleus quickly enough to avoid crystallization into graphite.

The spectra of quasars and NLS1 galaxies show surprising similarity in their spectral shape. They seem to scale only with the accretion rate. This is in contradiction with the simple expectations from the standard disk model which predicts lower disk temperature for higher black hole mass. Here we consider two mechanisms modifying the disk spectrum: the irradiation of the outer disk due to the scattering of the flux by the extended ionized medium (warm absorber and the development of the warm Comptonizing disk skin under the effect of the radiation pressure instability. Those two mechanisms seem to lead to a spectrum which indeed roughly scales, as observed, only with the accretion rate. The scenario applies only to objects with relatively high luminosity to the Eddington luminosity ratio for which disk evaporation is inefficient.

We analyze observations of the Seyfert galaxy NGC 4151 covering 90 years in the optical band and 27 years in the 2-10 keV X-ray band. We compute the Normalized Power Spectrum Density (NPSD), the Structure Function (SF) and the Autocorrelation Function (ACF) for these data. The results show that the optical and X-ray variability properties are significantly different. X-ray variations are predominantly in the timescale range of 5 - 1000 days. The optical variations have also a short timescale component which may be related to X-ray variability but the dominant effect is the long timescale variability, with timescales longer than $\sim$ 10 years. We compare our results with observations of NGC 5548 and Cyg X-1. We conclude that the long timescale variability may be caused by radiation pressure instability in the accretion disk, although the observed timescale in NGC 4151 is by a factor of few longer than expected. X-ray variability of this source is very similar to what is observed in Cyg X-1 but scaled with the mass of the black hole, which suggests that the radiation pressure instability does not affect considerably the X-ray production.

We derive power density spectra in the optical and X-ray band in the timescale range from several years down to $\sim $ a day. We suggest that the optical power density spectrum consists of two separate components: long timescale variations and short timescale variations, with the dividing timescale around 100 days. The shape of the short timescale component is similar to X-ray power density spectrum which is consistent with the interpretation of short timescale optical variations being caused by X-ray reprocessing. We show that the observed optical long timescale variability is consistent with thermal pulsations of the accretion disc.

We study the extensively monitored Seyfert galaxy NGC 5548 and compare its nuclear emission with models of accretion disk with accreting corona. To obtain the intrinsic nuclear spectra from the observed spectra we had to estimate and subtract the contribution from circumnuclear components such as stars, the Balmer continuum and blended FeII lines, and the FC2 extended, featureless continuum. The true nuclear spectra were compared with a two parameter model of the accreting disk with an accreting corona, described by the mass of the central black hole and viscosity. The model that best fitted the data was for M_{BH}=1.4e8 solar masses and the viscosity parameter alpha=0.033. Such a low viscosity parameter was necessary to produce the sufficient amount of X-rays. The vertical outflow of mass from corona in the form of wind had to be neglected in our model in order to fit into high and low states that NGC 5548 underwent. The model also predicts the behavior of the overall opt/UV/X continuum of NGC 5548 during the whole five year campaign.

We analyse the data from the optical/IUE observational campaign of the Seyfert galaxy NGC 5548 in the context of 10 phenomenological models. On the basis of the optical/UV data as well as constraints from the X-ray observations we can favour one model of the nucleus: an accretion disc with an inner radius cutoff surrounded by a hot corona. The second acceptable model for optical/UV data is a distribution of optically thin clouds. However, X-ray constraints which were crucial in the analysis of disc type models could not have been applied; further development of this model is necessary.