We present an updated and revised analysis of the relationship between the H \beta broad-line region ( BLR ) radius and the luminosity of the active galactic nucleus ( AGN ) .
Specifically , we have carried out two-dimensional surface brightness decompositions of the host galaxies of 9 new AGNs imaged with the Hubble Space Telescope Wide Field Camera 3 .
The surface brightness decompositions allow us to create “ AGN-free ” images of the galaxies , from which we measure the starlight contribution to the optical luminosity measured through the ground-based spectroscopic aperture .
We also incorporate 20 new reverberation-mapping measurements of the H \beta time lag , which is assumed to yield the average H \beta BLR radius .
The final sample includes 41 AGNs covering four orders of magnitude in luminosity .
The additions and updates incorporated here primarily affect the low-luminosity end of the R _ { BLR } – L relationship .
The best fit to the relationship using a Bayesian analysis finds a slope of \alpha = 0.533 ^ { +0.035 } _ { -0.033 } , consistent with previous work and with simple photoionization arguments .
Only two AGNs appear to be outliers from the relationship , but both of them have monitoring light curves that raise doubt regarding the accuracy of their reported time lags .
The scatter around the relationship is found to be 0.19 \pm 0.02 dex , but would be decreased to 0.13 dex by the removal of these two suspect measurements .
A large fraction of the remaining scatter in the relationship is likely due to the inaccurate distances to the AGN host galaxies .
Our results help support the possibility that the R _ { BLR } – L relationship could potentially be used to turn the BLRs of AGNs into standardizable candles .
This would allow the cosmological expansion of the Universe to be probed by a separate population of objects , and over a larger range of redshifts .