Microsimulation. Statistical methodology and assessment of uncertainty

Visa fullständig post

Permalänk

http://hdl.handle.net/10138/38871
Titel: Microsimulation. Statistical methodology and assessment of uncertainty
Författare: Zhou, M
Utgivare: Kela
Datum: 2013
Språk: Englanti
Tillhör serie: Working papers 44
ISSN: 2323-9239
Permanenta länken (URI): http://hdl.handle.net/10138/38871
Abstrakt: Nowadays, microsimulation method has been introduced to different fields, such as Social Science, Medicine research and Economic study. This method evaluates the effects of the proposed interventions or policies before they are implemented in the real world. In this article, I concentrate on microsimulation method used in Social Science by firstly explaining two main streams in microsimulation world, Static approach and Dynamic approach, in them, how statistical models are used are carefully explained by giving examples in Dynamic approach. In the following section, a Norwegian case is studied, this case gives the typical example of how the dynamic microsimulation used in the labor force and child care research, the effects of four different reform options are measured in this study. In the last section, the empirical study of a Finnish static microsimulation model (JUTTA) is carried out. The uncertainty of JUTTA is assessed and one of its sub model called Toimtuki (income-related supplementary benefit) is detected to have space to be more accurate. In order to do so, two statistical models - Linear Regression model and Two-Stage Least Squares (2SLS) model - are applied to it. From their results and diagnostics, we could conclude that both the Linear Regression and 2SLS successfully improves the accuracy of TOIMTUKI to some extent.
Beskrivning: 57 pages


Filer under denna titel

Totalt antal nerladdningar: Laddar...

Filer Storlek Format Granska
WorkingPapers44.pdf 661.7Kb PDF Granska/Öppna

Detta dokument registreras i samling:

Visa fullständig post