A Multiband Generalization of the Analysis of Variance Period Estimation Algorithm and the Effect of Inter-band Observing Cadence on Period Recovery Rate

We present a new method of extending the single band Analysis of Variance period estimation algorithm to multiple bands. We use SDSS Stripe 82 RR Lyrae to show that in the case of low number of observations per band and non-simultaneous observations, improvements in period recovery rates of up to $\approx$60\% are observed. We also investigate the effect of inter-band observing cadence on period recovery rates. We find that using non-simultaneous observation times between bands is ideal for the multiband method, and using simultaneous multiband data is only marginally better than using single band data. These results will be particularly useful in planning observing cadences for wide-field astronomical imaging surveys such as LSST. They also have the potential to improve the extraction of transient data from surveys with few ($\lesssim 30$) observations per band across several bands, such as the Dark Energy Survey.