Predicting Inflation Component Drivers in Nigeria: A Stacked Ensemble Approach

Abstract Our study examined the disaggregation of inflation components in Nigeria using the stacked ensemble approach, a machine learning algorithm capable of compensating the weakness of a base learner with the strength of another. This approach gives flexibility of a synergistic performance of stacking each base learner and produces a formidable model that yields the highest level of accuracy and best predictive ability. We analyzed the test data, out-of-sample, and our results show a strong accuracy in predicting inflation. Our results further show that food CPI is the most important driver for headline, urban, and rural inflation while bread and cereals is the most important driver for food inflation. However, biscuits, agric rice, garri white are among the top main drivers of bread and cereal inflation. We note that some CPI items that mostly drive inflation have lower weights while others have higher weights therefore, focusing entirely on CPI weights as a policy guide will stymied a successful control of inflation in Nigeria. In addition, ignoring CPI items with lower weights in policy intervention will make inflation difficult to control. Above all, adequate trace of the source of inflation to the least sub-component of each component will help address or formulates an appropriate policy to confront inflation problems in Nigeria. JEL: C53, E37