We discuss the construction of a photometric redshift catalogue ofluminous red galaxies (LRGs) from the Sloan Digital Sky Survey (SDSS),emphasizing the principal steps necessary for constructing such acatalogue: (i) photometrically selecting the sample, (ii) measuringphotometric redshifts and their error distributions, and (iii)estimating the true redshift distribution. We compare two photometricredshift algorithms for these data and find that they give comparableresults. Calibrating against the SDSS and SDSS-2dF (Two Degree Field)spectroscopic surveys, we find that the photometric redshift accuracy isσ~ 0.03 for redshifts less than 0.55 and worsens at higherredshift (~0.06 for z < 0.7). These errors are caused by photometricscatter, as well as systematic errors in the templates, filter curvesand photometric zero-points. We also parametrize the photometricredshift error distribution with a sum of Gaussians and use this modelto deconvolve the errors from the measured photometric redshiftdistribution to estimate the true redshift distribution. We pay specialattention to the stability of this deconvolution, regularizing themethod with a prior on the smoothness of the true redshift distribution.The methods that we develop are applicable to general photometricredshift surveys.

%B Monthly Notices of the Royal Astronomical Society %V 359 %P 237-250 %8 May 1, 2005 %G eng %U http://adsabs.harvard.edu/abs/2005MNRAS.359..237P %3 eprintid: arXiv:astro-ph/0407594