TERMIUM Plus^®

Fiche 1,  Justifications,  Anglais

Record number: 1, Textual support number: 1 CONT

The component scores for each day are computed and the 2873 by 6 data matrix are used in the clustering analysis. To determine the number of clusters that are [used] in the k-means procedure, the pseudo-F and pseudo-T2 statistics are used as guide lines (Davis and Kalkstein, 1990). The pseudo-F statistic describes the ratio of between cluster variance to within cluster variance. Peaks in the pseudo-F statistic are indicators of greater cluster separation. Pseudo-T2 quantifies the difference between the two clusters that are merged at a given step. Thus, if a given cluster (k) solution gives a jump in the pseudo-T2 statistic, than the k+1 cluster solution in selected as the optimal cluster. Any cluster represented with more than 20 percent of the total number of days would then be sub-clustered.  1, fiche 1, Anglais, - pseudo%2DF%20statistic

Fiche 1,  Justifications,  Français

Record number: 1, Textual support number: 1 CONT

À chaque permutation, une matrice de variables ajustées a été obtenue de par application de la procédure de régression [choisie]. Une statistique pseudo-F a été calculée pour chacune des matrices permutées [...]  1, fiche 1, Français, - statistique%20pseudo%2DF

Fiche 1,  Justifications,  Espagnol