| | 1 | = PDAF_diag_compute_moments = |
| | 2 | |
| | 3 | This page documents the routine `PDAF_diag_compute_moments` of PDAF, which was introduced with PDAF V3.0. |
| | 4 | |
| | 5 | This routine computes the mean, the unbiased variance, the unbiased skewness, and the unbiased excess kurtosis from an ensemble. All moments are stored in a single array. `kmax` controls the highest order that should be computed. |
| | 6 | |
| | 7 | The routine can be called in the pre/poststep routine of PDAF both before and after the analysis step to compute the ensemble statistics. |
| | 8 | |
| | 9 | The interface is the following: |
| | 10 | {{{ |
| | 11 | SUBROUTINE PDAF_diag_compute_moments(dim_p, dim_ens, ens, kmax, moments) |
| | 12 | }}} |
| | 13 | |
| | 14 | with the following arguments: |
| | 15 | |
| | 16 | - `dim_p` : INTEGER, INTENT(IN) [[BR]] Size of the state vector |
| | 17 | |
| | 18 | - `dim_ens` : INTEGER, INTENT(IN) [[BR]] Ensemble size |
| | 19 | |
| | 20 | - `ens` : REAL, DIMENSION(dim_p,dim_ens), INTENT(IN)[[BR]] Ensemble matrix |
| | 21 | |
| | 22 | - `kmax` : INTEGER, INTENT(IN) [[BR]] Highest order of statistical moment (1-4) |
| | 23 | |
| | 24 | - `moments` : REAL, DIMENSION(dim_p,kmax), INTENT(OUT) [[BR]] The statistical moments up to order `kmax` with |
| | 25 | - column 1: mean |
| | 26 | - column 2: variance |
| | 27 | - column 3: skewness |
| | 28 | - colmnu 4: excess kurtosis |