(C Venter, AK Harding, C Kalapotharakos, on behalf of the Fermi LAT Collaboration)
To date the Fermi Large Area Telescope (LAT) has detected at least 216 pulsars, many of which also pulsate in the radio band. This wealth of dual-band data has stimulated the development of sophisticated models of the pulsar magnetosphere, e.g., the force-free and force free inside dissipative outside (FIDO) models. Emission models based upon such magnetospheres yield concurrent radio and gamma-ray light curve predictions that can, in principle, constrain pulsar viewing geometries and electrodynamics via statistical goodness-of-fit testing. However, when concurrent fits are attempted via the Pearson's Chi-Squared test a clear bias toward the radio data emerges arising from the large disparity between the radio and gamma-ray relative light curve errors. To overcome this bias we have developed the Scaled-Flux Normalised Chi-Squared test statistic. The main difference between this and the traditional Chi-Squared test statistic is that it redefines the units in which the goodness-of-fit value for a single-band fit is expressed, allowing us to assign each concurrent dual-band fit a score between 0 and 1 with an associated Monte-Carlo-derived uncertainty. The goodness-of-fit values thus obtained can now be sensibly compared across both models and pulsars. As a first test of this new statistic we investigate the correlation between magnetospheric conductivity and pulsar spin-down luminosity uncovered by Brambilla et al. (2015) based on spectral fitting using the FIDO model. We will search for this trend by comparing predicted light curves from the FIDO-based model to observed radio and gamma-ray light curves of 16 young and millisecond pulsars.