The authors describe an application of neural networks to shortterm forecasting of mor tgage loan production in The Netherlands. Quite accurate out-of-sample pr edi c t ions are obtained using non-linear modelling. In the first par t some existing linear models are reviewed and an estimation of a mortgage production function is given in accord with the economic theory. Then an analysis of the neural network. structure, components and its training is provided. Finally the results obtained are tested and discussed.
Non-linear Modelling of Dutch mortgage Market / Gardin, F; Virili, Francesco. - In: ECONOMIC & FINANCIAL COMPUTING. - ISSN 0962-2780. - 5(2):(1995), pp. 131-145.
Non-linear Modelling of Dutch mortgage Market
VIRILI, Francesco
1995-01-01
Abstract
The authors describe an application of neural networks to shortterm forecasting of mor tgage loan production in The Netherlands. Quite accurate out-of-sample pr edi c t ions are obtained using non-linear modelling. In the first par t some existing linear models are reviewed and an estimation of a mortgage production function is given in accord with the economic theory. Then an analysis of the neural network. structure, components and its training is provided. Finally the results obtained are tested and discussed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
