Forecasting Inflation using Functional Time Series Analysis
Author: Raja Fawad Zafar

In present study we model the data using Functional Time Series Analysis (FTSA). The method is basically a univariate , so to check its performance, we compared it with seasonal Autoregressive Integrated Moving Average (ARIMA) models. We have used three data sets of monthly frequency from 2002-2011 to forecast Consumer Price Index CPI. General CPI, sector wise disaggregated CPI and City wise CPI. We withhold some data of last years (2011) and based on remaining year (2002-2010) we fitted model and forecasted the values of monthly CPI. Our study compares the performance of FTSA model and ARIMA model along with the performance of model based on all three types of data sets. The essence of the study is FTSA based on all three data sets performs better then Seasonal Autoregressive Integrated Moving Average (SARIMA) models. Comparison based on forecast evaluation criteria indicates that FTSA model using CPI general data outperforms all other models and comparison based on forecasted value for 2011, FTSA model based on sector wise disaggregated data outperforms all other models. Model wise comparison reveals the superiority of FTSA models and data wise comparison reveals better forecasting performance of sector wise disaggregated data. Supervisor:- Dr. Abdul Qayyum

Meta Data

Keywords : Functional Time Series Analysis, Inflation, Inflation Forcasting, Time Series Analysis
Supervisor: Abdul Qayyum

Related Thesis​