Single equation models for inflation forecasting in Rwanda

dc.contributor.authorKayisire, Pascal
dc.date.accessioned2014-07-22T13:02:07Z
dc.date.available2014-07-22T13:02:07Z
dc.date.issued2014-07-22
dc.description.abstractThis study evaluates Phillips curve forecasts of inflation for Rwanda. The study relies on the use of various single equation prototype Phillips curve models, as described by Stock and Watson (2008). Pseudo out-of-sample comparison tests are used to evaluate the forecast performance of these Phillips curve forecasts relative to the AR (autoregression) benchmark forecasts. In this regard, tests of equal forecast accuracy based on mean square forecast error and those based on forecast encompassing as used by several scholars (for example, Clark and McCracken (2001, 2005), Rapach and Weber (2004)) are reported. Furthermore, the results from forecasts using inflation in levels and in differences as the dependent variable are reported, to check the sensitivity to this specification issue. The study finds that the Phillips curve and augmented Phillips curve forecasts outperform the AR benchmark forecasts at one- and two-quarter horizons. The output gap, exchange rate and money supply (M3) are found to be good predictors of inflation in Rwanda in the generalised Phillips curve context. It is therefore strongly recommended that Rwandan economic policymakers take into consideration these variables when forecasting inflation.en_ZA
dc.identifier.urihttp://hdl.handle.net/10539/14966
dc.language.isoenen_ZA
dc.subject.lcshInflation (Finance) - Rwanda.
dc.titleSingle equation models for inflation forecasting in Rwandaen_ZA
dc.typeThesisen_ZA
Files
Original bundle
Now showing 1 - 2 of 2
No Thumbnail Available
Name:
P Kayisire Abstract.pdf
Size:
81.07 KB
Format:
Adobe Portable Document Format
No Thumbnail Available
Name:
P Kayisire Research Report.pdf
Size:
933.88 KB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Collections