Unemployment modelling covers both view of the labour market such as is, economy and knowledge of mathematics, statistics and, thus, econometrics. The importance of unemployment seems to be even more significant after the period of crisis; high unemployment is not only economic burden bud serious social risk and psychological problem as well. In the dissertation thesis, selected models used for unemployment modelling and -- in some cases for its prediction -- are introduced. To be able to predict the future trend of labour market reliably means to be able to plan tools of active and passive employment policies effectively. Alternatively, it means to search programs and supports that help in reduction of unemployment. Specific applications of models for the Czech labour market involve model of multistate life tables, simultaneous econometric models and Phillips curve. Phillips curve of mutual "trade-off" of unemployment and inflation is confirmed in short periods, in longer and long period of time rather fails, it is not reliable. It is not possible to use it for prediction at all; it would be needed to predict inflation. Analogous characteristics has the Beveridge curve. Simultaneous econometric models for number of economically active persons and for unemployment and inflation de facto fail, even though they demonstrate the range of opportunities including point and interval forecasts. Period of economic crisis when changes in labour market principles occur means usually problem for such the models, which work well in periods of stable growth or decline. More, it is difficult to specify these models correctly with regard to threat of multikolinearity. Multistate models aiming at calculation of multistate life tables, or even multistate projection are extremely demanding for input data. But they enable to understand relations or transitions among states, respectively. It is very beneficial tool for comprehension and policy planning in the area of labour market and social affairs in the process of lowering unemployment. Forecasts in such a type of model are possible but difficult because it is necessary to predict probability of transition among states.
| Identifer | oai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:201115 |
| Date | January 2010 |
| Creators | Miskolczi, Martina |
| Contributors | Langhamrová, Jitka, Fiala, Tomáš, Kotýnková, Magdalena |
| Publisher | Vysoká škola ekonomická v Praze |
| Source Sets | Czech ETDs |
| Language | Czech |
| Detected Language | English |
| Type | info:eu-repo/semantics/doctoralThesis |
| Rights | info:eu-repo/semantics/restrictedAccess |
Page generated in 0.0021 seconds