Skip to main navigation Skip to search Skip to main content

Modelling the CPI using a lognormal diffusion process and implications on forecasting inflation

  • Basel M. Al-Eideh*
  • , Ahmad S.A. Al-Refai
  • , Wafaa M. Sbeiti
  • *Corresponding author for this work
  • Economics Department
  • Kuwait University

Research output: Contribution to journal › Article › peer-review

7 Scopus citations

Abstract

The Maximum Likelihood estimator is used within a lognormal diffusion process and closed form analytical solutions are obtained. The monthly CPI forecasts are estimated for the period between 1970 and 2002. The quarterly estimates of inflation rates are obtained from monthly forecasts rather than from quarterly data. This has significantly improved the estimates of inflation rates. The model also produced a superior fit as compared to random walk and GARCH(p,q)-M models. The adopted approach is found to be simple, economical and generally suitable for modelling stochastic processes that reflect aggregation over time stemming from many factors, and in which the transition path between consecutive states is relatively smooth.

Original languageEnglish
Pages (from-to)39-51
Number of pages13
JournalIMA Journal of Management Mathematics
Volume15
Issue number1
DOIs
StatePublished - Jan 2004

Keywords

  • CPI, inflation
  • Lognormal diffusion
  • MLE
  • Stochastic differential equation

Fingerprint

Dive into the research topics of 'Modelling the CPI using a lognormal diffusion process and implications on forecasting inflation'. Together they form a unique fingerprint.

Cite this