TITLE:
Modeling Inflation in Bangladesh
AUTHORS:
Mohammad Lutfor Rahman, Mynul Islam, Manik Roy
KEYWORDS:
ARIMA, Inflation, Forecasting, Model Validity, Model Diagnostics, Bangladesh
JOURNAL NAME:
Open Journal of Statistics,
Vol.10 No.6,
December
10,
2020
ABSTRACT: Inflation has a substantial impact on an economy because it affects
the financial value of money and stability in the economy. Government and non-govern- ment policies might be hindered due to the excessive rate of inflation.
This paper aims to model and forecast inflation by the Box-Jenkins
autoregressive integrated moving average (ARIMA) technique using annual time
series data on inflation from 1987 to 2017 in Bangladesh. It is found that
ARIMA (2, 1, 0) model is the best optimal to forecast inflation for a period of
up to eight years. Graphical tools, as well as theoretical tests such as Ljung-Box, Shapiro-Wilk, and runs
tests have been used in the model diagnostics.