Non-normality, uncertainty and inflation forecasting: an analysis of china’s inflation

Economic forecasting is important because it can affect the decision making processes of individuals, firms and governments so as to affect their behaviours. In this thesis, I discuss different methodologies for forecasting and forecast evaluation. I also discuss the role of assumption of normality and the role of uncertainty in economic forecasting. The first chapter is the introduction of the thesis. In second chapter, I conduct a Monte Carlo simulation to investigate the performances of forecast combination and the forecast encompassing test under the forecast errors non-normality. In third chapter, I examines the relationship between inflation forecast uncertainties and macroeconomic uncertainty for China by using different measures of uncertainties. I also investigate the relationship between inflation forecast uncertainties and inflation itself. In fourth chapter, I compute the probabilities of deflation for China by applying density forecast based on the theories and methodologies from previous two chapters. Particularly, I construct density forecasts for different forecast horizons by a joint distribution using Student-t copula. The fifth chapter is conclusion.