Frequency of Jupiter analogues

http://exoplanets.astro.psu.edu/workshop/presentation/1-k-Tinney-AAPS.pdf

Quote :

Wittenmyer et al (2010, in prep)
- 3 Jupiter analogs from 123 stars with >8y data and >30 epochs, implies 3.6±1.4% at 3-6 AU.
- 120 non-detections implies upper limit of 37.2% for planets with K > 10 m/s in 3-6 AU

http://arxiv.org/abs/1011.4720

Quote :

The Anglo-Australian Planet Search has now accumulated 12 years of radial-velocity data with long-term instrumental precision better than 3 m/s. In this paper, we expand on earlier simulation work, to probe the frequency of near-circular, long-period gas-giant planets residing at orbital distances of 3-6 AU -- the so-called "Jupiter analogs." We present the first comprehensive analysis of the frequency of these objects based on radial-velocity data. We find that 3.3% of stars in our sample host Jupiter analogs; detailed, star-by-star simulations show that no more than 37% of stars host a giant planet between 3-6 AU.

IIRC that is fairly consistent with the microlensing results, bearing in mind that the orbital elements for such discoveries are usually not very well constrained.