Multivariate analysis of Côte d’Ivoire consumer price index: linear, regularized and generalized regression approaches

Abstract The high cost of living is a major challenge for Ivorian consumers. This paper seeks to explain this issue using Principal Component Analysis (PCA), Hierarchical Ascending Classification (HAC), as well as linear, regularized (Lasso, Ridge, Elastic Net), and generalized regression. The aim i...

Full description

Saved in:
Bibliographic Details
Main Authors: Aubin Yao N’Dri, Auguste Konan Kouakou, Amadou Kamagaté, Ouagnina Hili
Format: Article
Language:English
Published: Springer 2025-07-01
Series:Discover Analytics
Subjects:
Online Access:https://doi.org/10.1007/s44257-025-00036-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849238002782437376
author Aubin Yao N’Dri
Auguste Konan Kouakou
Amadou Kamagaté
Ouagnina Hili
author_facet Aubin Yao N’Dri
Auguste Konan Kouakou
Amadou Kamagaté
Ouagnina Hili
author_sort Aubin Yao N’Dri
collection DOAJ
description Abstract The high cost of living is a major challenge for Ivorian consumers. This paper seeks to explain this issue using Principal Component Analysis (PCA), Hierarchical Ascending Classification (HAC), as well as linear, regularized (Lasso, Ridge, Elastic Net), and generalized regression. The aim is also to identify the key determinants contributing to this phenomenon. Cluster analysis revealed three distinct periods characterized by different spending levels and economic and social priorities: Group 1: January 1997 to January 2003; Group 2: February 2003 to October 2012; Group 3: November 2012 to December 2023. Education emerged as the most discriminating factor, followed by closely related indices (FC, FHR, RH, MGS, CPI, HWE, Com, RC). These distinct periods particularly reflect changes in education policies, budget allocations to education, and variations in educational performance. Compared with Groups 1 and 3, the CPI increased by an average of 53.074%. This implies that the real value of 100000 FCFA is now equivalent to 65326.971 FCFA. An adjustment to restore purchasing power would require 153,076.13 FCFA. The study also revealed strong multicollinearity among the indices. The most significant variables influencing the CPI are FNAB, HWE, Tr, RH, MGS, and Edu, all of which are positively correlated with CPI. This suggests that an increase in any of these factors leads to a rise in CPI. Consequently, the high cost of living currently experienced by households is not solely driven by food expenditures but also by these broader economic variables.
format Article
id doaj-art-9b4e6b744da24a7e8a420bd7d53dcd37
institution Kabale University
issn 2731-8117
language English
publishDate 2025-07-01
publisher Springer
record_format Article
series Discover Analytics
spelling doaj-art-9b4e6b744da24a7e8a420bd7d53dcd372025-08-20T04:01:47ZengSpringerDiscover Analytics2731-81172025-07-013112110.1007/s44257-025-00036-3Multivariate analysis of Côte d’Ivoire consumer price index: linear, regularized and generalized regression approachesAubin Yao N’Dri0Auguste Konan Kouakou1Amadou Kamagaté2Ouagnina Hili3Environmental Sciences and Technologies Laboratory, University Jean LOROUGNON GUEDEDepartment of Economics and Management, University Jean LOROUGNON GUEDEEnvironmental Sciences and Technologies Laboratory, University Jean LOROUGNON GUEDEStatistics, Probability and Operations Research Laboratory, Félix Houphouët-Boigny National Polytechnic InstituteAbstract The high cost of living is a major challenge for Ivorian consumers. This paper seeks to explain this issue using Principal Component Analysis (PCA), Hierarchical Ascending Classification (HAC), as well as linear, regularized (Lasso, Ridge, Elastic Net), and generalized regression. The aim is also to identify the key determinants contributing to this phenomenon. Cluster analysis revealed three distinct periods characterized by different spending levels and economic and social priorities: Group 1: January 1997 to January 2003; Group 2: February 2003 to October 2012; Group 3: November 2012 to December 2023. Education emerged as the most discriminating factor, followed by closely related indices (FC, FHR, RH, MGS, CPI, HWE, Com, RC). These distinct periods particularly reflect changes in education policies, budget allocations to education, and variations in educational performance. Compared with Groups 1 and 3, the CPI increased by an average of 53.074%. This implies that the real value of 100000 FCFA is now equivalent to 65326.971 FCFA. An adjustment to restore purchasing power would require 153,076.13 FCFA. The study also revealed strong multicollinearity among the indices. The most significant variables influencing the CPI are FNAB, HWE, Tr, RH, MGS, and Edu, all of which are positively correlated with CPI. This suggests that an increase in any of these factors leads to a rise in CPI. Consequently, the high cost of living currently experienced by households is not solely driven by food expenditures but also by these broader economic variables.https://doi.org/10.1007/s44257-025-00036-3Consumer price indexHierarchical ascending classificationPrincipal component analysisLinearRegularized (Lasso, Ridge, Elastic Net)Generalized Regression
spellingShingle Aubin Yao N’Dri
Auguste Konan Kouakou
Amadou Kamagaté
Ouagnina Hili
Multivariate analysis of Côte d’Ivoire consumer price index: linear, regularized and generalized regression approaches
Discover Analytics
Consumer price index
Hierarchical ascending classification
Principal component analysis
Linear
Regularized (Lasso, Ridge, Elastic Net)
Generalized Regression
title Multivariate analysis of Côte d’Ivoire consumer price index: linear, regularized and generalized regression approaches
title_full Multivariate analysis of Côte d’Ivoire consumer price index: linear, regularized and generalized regression approaches
title_fullStr Multivariate analysis of Côte d’Ivoire consumer price index: linear, regularized and generalized regression approaches
title_full_unstemmed Multivariate analysis of Côte d’Ivoire consumer price index: linear, regularized and generalized regression approaches
title_short Multivariate analysis of Côte d’Ivoire consumer price index: linear, regularized and generalized regression approaches
title_sort multivariate analysis of cote d ivoire consumer price index linear regularized and generalized regression approaches
topic Consumer price index
Hierarchical ascending classification
Principal component analysis
Linear
Regularized (Lasso, Ridge, Elastic Net)
Generalized Regression
url https://doi.org/10.1007/s44257-025-00036-3
work_keys_str_mv AT aubinyaondri multivariateanalysisofcotedivoireconsumerpriceindexlinearregularizedandgeneralizedregressionapproaches
AT augustekonankouakou multivariateanalysisofcotedivoireconsumerpriceindexlinearregularizedandgeneralizedregressionapproaches
AT amadoukamagate multivariateanalysisofcotedivoireconsumerpriceindexlinearregularizedandgeneralizedregressionapproaches
AT ouagninahili multivariateanalysisofcotedivoireconsumerpriceindexlinearregularizedandgeneralizedregressionapproaches