Analyses of exoplanet statistics suggest a trend of giant planet occurrence with host star mass , a clue to how planets like Jupiter form .
One missing piece of the puzzle is the occurrence around late K dwarf stars ( masses of 0.5 - 0.75 M _ { \odot } and effective temperatures of 3900-4800 K ) .
We analyzed four years of Doppler radial velocities data of 110 late K dwarfs , one of which hosts two previously reported giant planets .
We estimate that 4.0 \pm 2.3 % of these stars have Saturn-mass or larger planets with orbital periods < 245 d , depending on the planet mass distribution and RV variability of stars without giant planets .
We also estimate that 0.7 \pm 0.5 % of similar stars observed by Kepler have giant planets .
This Kepler rate is significantly ( 99 % confidence ) lower than that derived from our Doppler survey , but the difference vanishes if only the single Doppler system ( HIP 57274 ) with completely resolved orbits is considered .
The difference could also be explained by the exclusion of close binaries ( without giant planets ) from the Doppler but not Kepler surveys , the effect of long-period companions and stellar noise on the Doppler data , or an intrinsic difference between the two populations .
Our estimates for late K dwarfs bridge those for solar-type stars and M dwarfs and support a positive trend with stellar mass .
Small sample size precludes statements about finer structure , e.g .
a “ shoulder ” in the distribution of giant planets with stellar mass .
Future surveys such as the Next Generation Transit Survey and the Transiting Exoplanet Satellite Survey will ameliorate this deficiency .