We take a fresh look at the determination of distances and velocities of neutron stars. The conversion of a parallax measurement into a distance, or distance probability distribution, has led to a debate quite similar to the one involving Cepheids, centering on the question whether priors can be used when discussing a single system. With the example of PSRJ0218+4232 we show that a prior is necessary to determine the probability distribution for the distance. The distance of this pulsar implies a gamma-ray luminosity larger than 10% of its spindown luminosity. For velocities the debate is whether a single Maxwellian describes the distribution for young pulsars. By limiting our discussion to accurate (VLBI) measurements we argue that a description with two Maxwellians, with distribution parameters sigma1=77 and sigma2=320 km/s, is significantly better. Corrections for galactic rotation, to derive velocities with respect to the local standards of rest, are insignificant.

We argue that comparison with observations of theoretical models for the velocity distribution of pulsars must be done directly with the observed quantities, i.e. parallax and the two components of proper motion. We develop a formalism to do so, and apply it to pulsars with accurate VLBI measurements. We find that a distribution with two maxwellians improves significantly on a single maxwellian. The `mixed' model takes into account that pulsars move away from their place of birth, a narrow region around the galactic plane. The best model has 42% of the pulsars in a maxwellian with average velocity sigma sqrt{8/pi}=120 km/s, and 58% in a maxwellian with average velocity 540 km/s. About 5% of the pulsars has a velocity at birth less than 60\,km/s. For the youngest pulsars (tau_c<10 Myr), these numbers are 32% with 130 km/s, 68% with 520 km/s, and 3%, with appreciable uncertainties.

We use a Bayesian approach to derive the distance probability distribution for one object from its parallax with measurement uncertainty for two spatial distribution priors, viz. a homogeneous spherical distribution and a galactocentric distribution - applicable for radio pulsars - observed from Earth. We investigate the dependence on measurement uncertainty, and show that a parallax measurement can underestimate or overestimate the actual distance, depending on the spatial distribution prior. We derive the probability distributions for distance and luminosity combined, and for each separately, when a flux with measurement error for the object is also available, and demonstrate the necessity of and dependence on the luminosity function prior. We apply this to estimate the distance and the radio and gamma-ray luminosities of PSR J0218+4232. The use of realistic priors improves the quality of the estimates for distance and luminosity, compared to those based on measurement only. Use of a wrong prior, for example a homogeneous spatial distribution without upper bound, may lead to very wrong results.