Using nonstationary time series for estimating small open economy model with financial frictions
| Authors | |
|---|---|
| Year of publication | 2012 |
| Type | Article in Proceedings |
| Conference | Proceedings of 30th International Conference Mathematical Methods in Economics |
| MU Faculty or unit | |
| Citation | |
| Web | Conference paper |
| Field | Economy |
| Keywords | DSGE; nonstationary; balanced growth path; filter |
| Description | This paper compares results of small open economy DSGE model estimation with prefiltered data and non-prefiltered data. There are at least two ways of taking a model to data: (i) filtering the historical time series outside the model in order to render them stationary or (ii) solving the model around balanced growth path and using nonstationary time series in estimation. While filtering is ubiquitous, there are number of problems associated with it. In particular, prefiltering time series outside of the model using univariate filters (the usual method) results in loss of information. This paper employs small open economy model with financial accelerator to show how prefiltering using univariate filters influences estimates of model parameters and the output gap. It concludes that for small-scale small open economy DSGE models, other ways of dealing with the filtering problem might be worth considering. |