Appendix A: SXLF of the `type 1' AGN sample
In the main part of this paper, we have concentrated on the SXLF expression for the mixture of type 1 and type 2 AGNs for the reasons explained in Sect. 3.1. However, since previous works in literature mainly give expression for only type 1 AGNs (with broad permitted lines), it is of significant historical interest to investigate the SXLF properties for only type 1 AGNs. Because our samples come from several different sources and every subsample has its own criteria for classifying AGNs into subclasses, our expressions given here should not be used for any quantitative work (e.g. using it as a starting point of a population synthesis modeling under an assumption that they represent the unabsorbed AGNs) without assessment of possible biases described in Sect. 3.1.
We have defined the `type 1' AGN sample as follows. We have included AGNs explicitly classified in the original catalogs as Seyfert 1-1.5's, BLRGs, and QSOs, while excluded those classified as Seyfert 1.8-2, NELGs, and Narrow-line Seyfert 1's (NLS1). The NLS1's have been excluded since they would not have been included in the `broad-line' AGN samples in the previous works, especially those with low-quality optical spectra. A number of RBS objects classifed simply as `AGNs' have been ckecked with the NED database and/or the original spectra for the subclassifications. For the Lockman Hole sample, we have included objects with ID classes (a)-(c) (see Schmidt et al. 1998) and excluded (d)-(e). For the Marano sample, those classified as AGNs in Zamorani et al. (1999) have been assumed to be type 1 AGNs unless otherwise stated, since type 2 AGNs have been explicitly noted. The AGNs which have not been subclassified using the above procedure have been excluded from our `type 1' sample. The fraction of AGNs included in this `type 1' sample are 98% (RBS), 90% (SA-N), 93%(RIXOS), 88% (NEP+Marano+UKD), and 85% (LH) respectively.
We have only considered the PLE and LDDE1 models in this appendix. Table A1 shows best-fit parameters and KS probabilities (see main text for details). Table A1 shows that PLE is still rejected with a large significance for both cosmologies, while finding good fits with the LDDE1 form. A plot similar to Fig. 9 is shown for the `type 1' AGN sample in Fig. A1.
© European Southern Observatory (ESO) 2000
Online publication: December 8, 1999