Artikel Jurnal Internasional Melengkapi survei inflasi Afrika Selatan: alat peramalan yang sesuai
The central bank is currently conducting a survey of inflation expectations to better align inflation expectations with expectations of the general public. However, surveys are time-consuming, complicated, expensive and not always accurate, so sacrificing this credibility is hope. The complexity of inflation targeting and forecasting difficulties in real time can also cause policymakers to consider more basic models, which can lead to inaccurate estimates. This article uses a less complicated model, such as the seasonally adjusted integrated moving average autoregressive and Holt-Winters exponential smoothing model,
to provide equally reliable estimates. More complex approaches in non-linear form
The process of autoregressive neural networks is also used to model strategy and rationalism [ways] in which the general public formulates their expectations. Overall, the forecast   estimates provided by these models are superior when compared to inflation expectations provided by the International Monetary Fund, South Africa Reserve Bank and
Bureau of Economic Research.