Multivariate and predictive analysis of Côte d'Ivoire consumer price index

The high cost of living is a problem facing consumers in Cˆote d’Ivoire. This paper attempts to explain this high cost of living using principal component analysis (PCA) and hierarchical ascending classification (HAC). The aim is also to achieve the best possible prediction of the consumer price index (CPI) using multiple linear regression, and then to find the determinants involved in its explanation. The PCA and HAC methods have revealed three clusters. Cluster1: January 1997 to January 2003; cluster2: February 2003 to October 2012 and cluster3: November 2012 to December 2023. Compared with clusters 1 and 3, the CPI increased by an average of 53.074%. Consequently, the real value of 100000 FCFA is now 65326.971 FCFA. The high cost of living began in February 2003. It has intensified since December 2012. The relationships between the indices were studied. It was concluded that there is a strong correlation and also multicollinearity between them. A model was built to predict the CPI. In long term, the indices to be taken into account are the Housing, Water, Electricity, Gas and Other Fuels (HWE) and, Food and non-alcoholic beverages. In short term, the HWE and, Restaurants and Hotels indices. An interpretable model with the sub-indices was constructed using principal component regression. It revealed that all the sub-indices are positively correlated with the CPI. This means that a positive variation in one of them will lead to an increase in this index. MSC Classification: 62A09 , 62H25 , 62H30 , 62J10 , 62P12