Salīdzināt metodes

	Latent Class Analysis (LCA)	Lineārā diskriminantā analīze
Nozare	Statistika	Statistika
Saime	Latent structure	Latent structure
Izcelsmes gads	1950s–1968	1936
Autors	Paul F. Lazarsfeld	Ronald A. Fisher
Tips	Latent variable / person-centered classification	Supervised classification and dimension reduction
Pirmavots	Goodman, L. A. (1974). Exploratory latent structure analysis using both identifiable and unidentifiable models. Biometrika, 61(2), 215–231.	Fisher, R. A. (1936). The use of multiple measurements in taxonomic problems. Annals of Eugenics, 7(2), 179–188.
Citi nosaukumi	LCA, latent class model, latent categorical analysis, finite mixture of multinomials	LDA, Fisher discriminant analysis, discriminant function analysis, canonical discriminant analysis
Saistītās	6	4
Kopsavilkums	Latent class analysis identifies unobserved subgroups — latent classes — within a population by finding patterns of responses across a set of categorical observed indicators. It is the categorical-variable counterpart of cluster analysis, but grounded in an explicit probabilistic model, and is widely used in social, health, and behavioral sciences to discover typologies in survey or diagnostic data.	Discriminant analysis finds linear combinations of predictor variables that best separate two or more known groups. It is used both to understand which predictors distinguish the groups and to classify new observations into those groups with minimum error.