A Bayesian analysis of 47 Ursae Majoris ( 47 UMa ) radial velocity data confirms and refines the properties of two previously reported planets with periods of 1079 and 2325 days . The analysis also provides orbital constraints on an additional long period planet with a period \sim 10000 days . The three planet model is found to be ~ { } 10 ^ { 5 } times more probable than the next most probable model which is a two planet model . The nonlinear model fitting is accomplished with a new hybrid Markov chain Monte Carlo ( HMCMC ) algorithm which incorporates parallel tempering , simulated annealing and genetic crossover operations . Each of these features facilitate the detection of a global minimum in \chi ^ { 2 } . By combining all three , the HMCMC greatly increases the probability of realizing this goal . When applied to the Kepler problem it acts as a powerful multi-planet Kepler periodogram . The measured periods are 1078 \pm 2 , 2391 _ { -87 } ^ { +100 } , and 14002 _ { -5095 } ^ { +4018 } d , and the corresponding eccentricities are 0.032 \pm 0.014 , 0.098 _ { - .096 } ^ { + .047 } , and 0.16 _ { - .16 } ^ { + .09 } . The results favor low eccentricity orbits for all three . Assuming the three signals ( each one consistent with a Keplerian orbit ) are caused by planets , the corresponding limits on planetary mass ( M \sin i ) and semi-major axis are ( 2.53 _ { - .06 } ^ { + .07 } M _ { J } , 2.10 \pm 0.02 au ) , ( 0.54 \pm 0.07 M _ { J } , 3.6 \pm 0.1 au ) , and ( 1.6 _ { -0.5 } ^ { +0.3 } M _ { J } , 11.6 _ { -2.9 } ^ { +2.1 } au ) , respectively . Based on a three planet model , the remaining unaccounted for noise ( stellar jitter ) is 5.7 m s ^ { -1 } . The velocities of model fit residuals were randomized in multiple trials and processed using a one planet version of the HMCMC Kepler periodogram . In this situation periodogram peaks are purely the result of the effective noise . The orbits corresponding to these noise induced periodogram peaks exhibit a well defined strong statistical bias towards high eccentricity . We have characterized this eccentricity bias and designed a correction filter that can be used as an alternate prior for eccentricity , to enhance the detection of planetary orbits of low or moderate eccentricity .