Forecasting Inflation in Pakistan



Forecasting Inflation in Pakistan

Introduction

Inflation is known as a persistent rise in prices of products and services that might further be related to an increase in the volume of money, thus a decrease in the value of currency. Through inflation, the economy of entire nations can become affected; especially for multinational corporations operating within a country the trends in currency can be highly significant. Inflation is sometimes caused by the increase in cost of businesses. With increment in the cost of production of goods and services, the final price of the product will go up, resulting in a domino effect for the rest of the business operations. The loss in revenue would affect future operations and employers would have to increase pay of employees and so on. Dramatic fluctuations in the global economic environment have been noticed during the past three decades, the result of which has been several studies regarding inflation models and forecasts. Much work has been done on the examination and evaluation of inflation forecasts as well as methodologies that may be put to use in said forecasts. In literature studies, Fama (1975, 1977) has been well-known for his theories, some of which have been extended by Gibbons (1982, 1984). These theories present approaches that aim to extract the interest rates circulating within the market. This is done with the inherent forecast of inflation, thus creating time-series models to form an opinion of what the interest rates may be. In this paper, I will be examining inflation rates in Pakistan, a developing nation with inconsistent currency trends. The level of forecasting will be examined in Pakistan and its resulting inflation.

Method

The variable to be modeled in this research is inflation rate. The univariate model as well as the time series models has been tested among theorists with the conclusion that the latter has less of a margin for error. In calculating nominal interest rates during inflationary periods in economy, the utilization of a time series model can form forecasts with greater accuracy. Further improvement of the time series model came forth through Aiden Meyler, Geoff Kenny and Terry Quinn in 1998, with the ARIMA models. These models were a variation on the time series models, and would go on to forecast the inflation in Ireland at the time. Additionally, the Bayesian approach was introduced in order to estimate the vector auto aggressive models. These were a further improvement in the estimation of forecasts.

Forecasting Model

In this research, we will incorporate percentage change model to forecast inflation. The equation that will be used to estimate the percentage change in inflation over the years is illustrated as:

Percent change = [(Vpresent-Vpast)/Vpast] * 100

Based on the estimated change in inflation over the period of ten years, we will predict inflation for the period of 2013:

Years

Inflation Rate

2002

2.504

2003

3.102

2004

4.568

2005

9.276

2006

7.921

2007

7.771

2008

11.998

2009

20.775

2010

11.5

2011

7.5

2012

6

By incorporating percentage change formula, we have calculated average percentage change in inflation during 10 years. The average change is -33%. By taking in to account this percentage change, we will calculate inflation for the ...