Modeling Consumer Price Index (CPI) Inflation Using Seasonal Auto-Regressive Integrated Moving Average (SARIMA) METHOD: an Empirical Consumer Price Index (CPI) Study of Pakistan 2014-2024
Abstrak
Inflation is a critical macroeconomic variable that directly impacts economic stability and policymaking. Accurate forecasting of inflation is essential for effective monetary policy, business planning, and investment decision-making. This study aims to model and forecast inflation in Pakistan using the Seasonal AutoRegressive Integrated Moving Average (SARIMA) method, with a specific focus on the Consumer Price Index (CPI) as a key measure of inflation. Utilizing CPI data from 2014 to 2024, the research seeks to identify the most suitable SARIMA model to capture both seasonal and non-seasonal patterns within the data. The SARIMA model, an extension of the ARIMA framework, integrates seasonal autoregressive, moving average, and differencing components to address the complexities inherent in time series data. The study employs rigorous model selection criteria, including the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC), alongside diagnostic checks to ensure robust forecasting. Both in-sample and out-of-sample validation will be conducted to evaluate the predictive accuracy of the selected model. The outcomes of this research are expected to provide actionable insights for policymakers, particularly the State Bank of Pakistan and other relevant stakeholders, in formulating evidence-based strategies to manage inflation. Additionally, the findings will serve as a valuable resource for households and businesses in planning their savings and investments. By offering a reliable methodology for inflation forecasting, this study contributes to the broader understanding of macroeconomic stability in Pakistan.