Indexing

By default, CINDERELLA expects the file index of time series and SIGSPEC results to start at zero. If the user wants the program to start at a different index, the keyword mfstart may be given in the .cnd file. This keyword is followed by an integer representing the desired start index. Furthermore, the software takes into account as many files with consecutive indices as available. The number of indices to process may be restricted by means of the keyword multifile, followed by an integer representing the last index to take into account. The use of the keywords mfstart and multifile is strictly consistent with the SIGSPEC conventions (SIGSPEC manual, p.).

Example. The sample project index contains the same input as the project CinderellaNative (p.), but both the time series input files and the SIGSPEC result files are now enumerated from 004847 to 004855 rather than from 000000 to 000008. The file index.cnd consists of a single line,

mfstart 4847

Since the keyword multifile is not specified in the file index.cnd, CINDERELLA uses all input files available through consecutive indices.

The computations are entirely the same as for CinderellaNative, but all indices are consistently incorporated by CINDERELLA. E.g., the screen output for the single-comparison computations is now:

*** pairwise Cinderella analysis ***************************

  4848 vs.   4847: conditional sig
  4849 vs.   4847: conditional sig
  4850 vs.   4847: conditional sig
  4851 vs.   4847: conditional sig
  4852 vs.   4847: conditional sig
  4853 vs.   4847: conditional sig
  4854 vs.   4847: conditional sig
  4855 vs.   4847: conditional sig