Šiame darbe buvo nagrinėjamas dinaminis stochastinis stacionarus modelis aprašytas I.Carabenciov, I.Ermolaev, Ch.Freedman, M.Juillard, O.Kamenik, D.Korshunov, D.Laxton „A Small Quarterly Projection Model of the US Economy“ straipsnyje. Duomenis pateikė „Euromonitor International“ įmonė. Modelyje naudojami keturi Jungtinių Valstijų ekonomikos rodikliai: realus bendras vidaus produktas, nedarbo lygis, infliacija ir federalinių fondų palūkanų norma. Rodikliai stebimi 1994 m. І ketv. – 2009 m. ІІ ketv. laikotarpiu. Darbo tikslas – pakartoti „A Small Quarterly Projection Model of the US Economy“ straipsnio rezultatus. Buvo pasirinkti du programavimo paketai – Dynare ir Winbugs. Modelis buvo suprogramuotas dvejomis skirtingomis programomis, kurios remiasi Bajeso metodologija. Po to, gauti rezultatai buvo lyginami su straipsnyje pateiktais rezultatais. Atlikus visus skaičiavimus buvo gauti tokie rezultatai: su Dynare puikiai pavyko pakartoti modelio rezultatus. Su Winbugs programa gauti rezultatai nepilnai sutapo su straipsnyje pateiktais rezultatais. Iš to galima buvo padaryti išvada, kad Dynare programa labiau tinka dinaminių stochastinių stacionarių modelių vertinimui. Šio darbo rezultatai bus labai naudingi tyrimo planuotojams bei vykdytojams. Modelio rezultatai bus įtraukti į globalųjų modelį, kuris apjungs dar kelių valstybių modelius. / In the research paper a dynamic stochastic general equilibrium described in I.Carabenciov, I.Ermolaev, Ch.Freedman, M.Juillard, O.Kamenik, D.Korshunov, D.Laxton article „A Small Quarterly Projection Model of the US Economy“ was analyzed. The data for analysis was provided by “Euromonitor International“ company. The benchmark model has only four variables: real gross domestic product (GDP), unemployment rate, consumer price index, federal funds rate. The model is estimated over sample period from 1994QI till 2009QII. The aim of the research was to reiterate the results given in the article „A Small Quarterly Projection Model of the US Economy“. For this aim two programming packages were chosen – Dynare and Winbugs. The model was programmed using two different programs, both based on Bayesian methodology. Afterward the results were compared with results presented in the article. After all the calculations were done, the results were following: model was successfully repeated with Dynare. Although the results obtained with Winbugs program differed from the results given in the article. It therefore concluded that Dynare program is more suitable for the assessment of stationary stochastic dynamic models.
| Identifer | oai:union.ndltd.org:LABT_ETD/oai:elaba.lt:LT-eLABa-0001:E.02~2011~D_20140627_171859-84042 |
| Date | 27 June 2014 |
| Creators | Ulanovska, Anastazja |
| Contributors | Eidukevičius, Rimantas, Vilnius University |
| Publisher | Lithuanian Academic Libraries Network (LABT), Vilnius University |
| Source Sets | Lithuanian ETD submission system |
| Language | Lithuanian |
| Detected Language | English |
| Type | Master thesis |
| Format | application/pdf |
| Source | http://vddb.library.lt/obj/LT-eLABa-0001:E.02~2011~D_20140627_171859-84042 |
| Rights | Unrestricted |
Page generated in 0.0023 seconds