Title: Big data forecasting of South African inflation
Authors: Kevin Kotze - University of Cape Town (South Africa) [presenting]
Abstract: The use of statistical learning techniques and big data to enhance the accuracy of inflation forecasts is investigated. We make use of a large dataset for the disaggregated prices of consumption goods and services, which we partially reconstruct, and a large suite of different statistical learning and traditional time series models. We find that the statistical learning models are able to compete with most benchmarks over medium to longer horizons, despite the fact that we only have a relatively small sample of available data, but are usually inferior over shorter horizons. Our findings suggest that this result may be attributed to the ability of these models to make use of relevant information, when it is available, and may be particularly useful during periods of crisis, when deviations from the steady state are more persistent. We find that the accuracy of the central bank's near-term inflation forecasts compares favourably with those of other models, while the inclusion of off-model information, such as electricity tariff adjustments and other sources of within-month data, provides these models with a competitive advantage. Lastly, we generate Shapley values for selected statistical learning models to identify the most important contributors to future inflationary pressure and also investigate the relative performance of the different models as we experienced the effects of the pandemic.