oph_aggregate_stats_partial¶

Description¶

Behaviour

It computes partial results useful to compute a statistical index based on mask value working vertically on the columns of arrays of measures. The partial results, after a merge, can be treated as input to oph_aggregate_stats_final primitive to compute the final results for the requested indexes. Possible partial results, depending on the requested indexes, are:

• sum{i}: number of operands;
• sum{x_i}: sum of operands;
• sum{x_i²}: sum of squared operands;
• sum{x_i³}: sum of operands raised to 3;
• sum{x_i⁴}: sum of operands raised to 4;
• max{x_i}: maximum value between operands;
• min{x_i}: minimum value between operands.

For example, in case the 3-order moment (skew) is requested, the output will contain sum{i}, sum{x_i}, sum{x_i²} and sum{x_i³}.

Parameters

• input measure type: Ophidia typing. Supported types are: ‘oph_double’; ‘oph_float’; ‘oph_long’; ‘oph_int’; ‘oph_short’; ‘oph_byte’.
• output measure type: Ophidia typing. Supported types are: ‘oph_double’; ‘oph_float’; ‘oph_long’; ‘oph_int’; ‘oph_short’; ‘oph_byte’.
• measure: input measure.
• mask: 7-bit mask
  • 0000000: invalid
  • 1xxxxxx: mean
  • x1xxxxx: variance
  • xx1xxxx: std dev
  • xxx1xxx: skew
  • xxxx1xx: excess kurtosis
  • xxxxx1x: max
  • xxxxxx1: min
  • 1111111: all (default)
  • 101: mean + std dev
  • ...

Return type

Binary-array.

Examples

Compute partial results for mean, maximum and minimum of a set of time series.

oph_aggregate_stats_partial('OPH_DOUBLE','OPH_DOUBLE',measure,'1000011')

Operation type

Aggregate.

Arguments¶