Subsample and jacknife: A general technique for the estimation of sampling errors, with applications and examples in the field of transport planning

A method of estimating the effect of sampling error on derived results is described. The technique involves the division of the full sample into overlapping subsamples and the subsequent parallel processing of these subsamples. The particular advantage of the technique is the way in which it can be used to follow automatically the effect of sampling variance through the most complex operations, such as the calibration of a transport model.