This article identifies the best models for forecasting the volatility of daily exchange returns of developing countries. An emerging consensus in the recent literature focusing on industrialized countries has noted the superior performance of the Fractionally Integrated Generalized Autoregressive Conditionally Heteroscedastic (FIGARCH) model in the case of industrialized countries, a result that is reaffirmed here. However, we show that when dealing with developing countries' data the IGARCH model results in substantial gains in terms of the in-sample results and out-of-sample forecasting performance.
|Pages (from-to)||1675 - 1691|
|Journal||Applied Financial Economics|
|Publication status||Published - 1 Oct 2013|
Austrian Classification of Fields of Science and Technology (ÖFOS)
- 502010 Public finance