New periodic variables from the Hipparcos epoch photometry Two selection statistics are used to extract new candidate periodicvariables from the epoch photometry of the Hipparcos catalogue. Theprimary selection criterion is a signal-to-noise ratio. The dependenceof this statistic on the number of observations is calibrated usingabout 30000 randomly permuted Hipparcos data sets. A significance levelof 0.1 per cent is used to extract a first batch of candidate variables.The second criterion requires that the optimal frequency be unaffectedif the data are de-trended by low-order polynomials. We find 2675 newcandidate periodic variables, of which the majority (2082) are from theHipparcos`unsolved' variables. Potential problems with theinterpretation of the data (e.g. aliasing) are discussed.
