Three ongoing microlensing experiments have found more candidate events than expected from the known stars. These experiments measure only one parameter of the massive compact halo objects (machos), the magnification time scale of the events. More information is required to understand the nature of the machos. A satellite experiment has been proposed to measure their projected transverse speed $\tilde{v} = v/(1-z)$, where $v$ is the macho transverse speed and $z$ its distance divided by the distance of the source. Measurement of $\tilde{v}$ would determine whether the machos were in the Galactic disk, Galactic halo, or in the Large Magellanic Cloud (LMC). We simulate events observed toward the LMC by the Earth and by a satellite in an Earth like heliocentric orbit. To leading order, such an experiment determines $\tilde{v}$ up to a two fold degeneracy. More precise measurements break the degeneracy. We show that with photometric precisions of 3\% to 4\% and approximately 1 observation per day, $\tilde{v}$ can be measured with a maximum error of 20\% for 70\% to 90\% of events similar to the ones reported by the EROS and MACHO collaborations. The projected transverse velocity is known with the same maximum error for 60\% to 75\% of these events. This 20\% maximum error is not a 1 $\sigma$ error but is mostly due to degeneracy between two possible solutions, each one being localized to much better than 20\%. These results are obtained with an Earth-satellite separation of 1 AU, and are improved by a larger separation.