Method for measuring of latent indicators of continuous sets of original information data

Autor: S. I. Moiseev, R. V. Kuzmenko, L. V. Stepanov
Rok vydání: 2017
Předmět:
Zdroj: 2017 2nd International Ural Conference on Measurements (UralCon).
Popis: This paper proposes an estimation method for the latent variable Rasch model based on the method of least squares which allows a continuous data set using. The research suggests the application of original approaches within the method for the solution of some applied problems. The authors explain how to use it for task assignment and work organization, decision-making under certainty and the securities portfolio formation. The conducted computational experiments showed that the model based on the method of least squares is characterized by high statistical accuracy and can be used to calculate the latent variable estimates. At the same time, this model has a much wider range of practical applications than the classical Rasch model. In compliance with the Rasch model based on the method of least squares the authors proposed a new approach to the work organization, methods of decision making under certainty, models of expert estimation by the method of hierarchies analysis, the algorithm of securities portfolio formation. The model is based on the method of least squares can also be used to estimate the latent indicators during the evaluation of quality, efficiency, and other areas of theoretical and practical studies. In particular, there are works in which the described model is used in education, expert evaluation as well as for the assessment of software quality and personnel competence.
Databáze: OpenAIRE