Movatterモバイル変換

Changes in version 0.3.1

By default, the data is now centered and scaled before change pointdetection is applied, which improves empirical performance. This can beturned off by setting the new argumentscale.data = FALSEin the functionsnp.mojo(),np.mojo.multilag(), andmultiscale.np.mojo().

The associated paper is now accepted for publication in Biometrika:seedoi:10.1093/biomet/asaf024 for full details.
Updatedmultiscale.np.mojo() function so that returnedcpts are given in time order.
The p-values returned bynp.mojo(),np.mojo.multilag(), andmulsticale.np.mojo()have been replaced by importance scores: when the bootstrap is used,these are essentially one minus the p-value. Larger scores give moreevidence of a change point.
Fixed error that occurs when using the manual threshold and thenp.mojo.multilag() function, thanks to Chuanyang Zhang forspotting this.
You can now use different manual thresholds for different lags: foruse innp.mojo.multilag(), you can supply a vector, whilstformulsticale.np.mojo(), you can supply a list ofvectors.
Added package level documentation: see?CptNonPar.

Added a new functionmultiscale.np.mojo(), thisfunction performs multiscale, multi-lag nonparametric change pointdetection using a set of bandwidths (as opposed to a single bandwidth asinnp.mojo.mulilag()).
Updated parallelisation code innp.mojo() to avoidunnecessary call tocloseAllConnections(), thanks to HenrikBengtsson for highlighting this.
Changed uses of itemize in .Rd files to comply with CRAN check.