Моделирање со структурни равенки и примена

Vitanova, Vasilka (2014) Моделирање со структурни равенки и примена. Masters thesis, Goce Delcev University, Stip.


Download (2788Kb) | Preview


Statistics is a tool that is essential in many scientific disciplines, particularly in areas where there is a need to process and analyze large amounts of data. Today's era of modern technology allows everyone to relatively easy assemble and work with large amounts of data. Such data is difficult to analyze with classical statistical techniques due to the large number of variables and the nature of the interrelation between these variables. Multivariate statistical methods are used to analyze these data in multiple sciences. SEM (structural equation modeling-modeling with structural equations) is a statistical technique that combines elements of traditional multivariate modeling, regression analysis, factor analysis and the results of both these approaches are used as coefficients in the definition of systems of linear equations (equation modeling ). [53] Structural equation modeling or SEM is relatively general, and is mainly linear and intersection of several techniques of statistical modeling. The factor analysis, path analysis and regression analysis are special cases of SEM. SEM is more of a confirmation than a research technique. This means that the researcher is more likely to use SEM to determine whether a model is valid, rather than use SEM to "find" a suitable model, although SEM analyses often involve certain research element. Compared with the factor and regression analysis, SEM is relatively unknown technique and application, and has roots in the works that appear in the late 60s of the 20th century. As such methodology is still in development, and even basic concepts are subject to challenge and revision.

Item Type: Thesis (Masters)
Subjects: Natural sciences > Computer and information sciences
Divisions: Faculty of Computer Science
Depositing User: Dimitar Ljubotenski
Date Deposited: 03 Jul 2014 13:53
Last Modified: 03 Jul 2014 13:54
URI: http://eprints.ugd.edu.mk/id/eprint/10396

Actions (login required)

View Item View Item