Inflation is a persistent rise in the general pricelevels of goods and services in an economy over a period of time. Inflation isgenerally measured in terms of a consumer price index (CPI), which tracks theprices of a basket of core goods and services over time.
Inflation has been oneof the most economic challenges in the world; mostly in developing countries(Omekara 2013).Inflation takes place when there is a sustained increase in thegeneral price level. Traditionally high inflation rates are considered to bedamaging an economy. High inflation creates uncertainty and can dry away thevalue of savings. Although, most Central Banks target an inflation rate of 2%,proposing that low inflation can have various advantages to the economy (Lenco2014). A fall in prices (negative inflation) is very dangerous during anextended period of deflationary pressures. When prices are falling people areunwilling to spend money because they are worried that price will be cheaper inthe future, therefore, they keep delaying purchases. Also a fall in priceincreases the real value of debt and reduces disposable income of individualswho are struggling to pay their debt.
When a debt is taken as a mortgage, theynormally anticipate an inflation rate of 2% to help erode the value of debtover time. Tanzania has been facing inflation since acquire of independence in1961 which is 56 years ago. According to National Bureau of Statistics (NBS),Tanzania’s inflation rate historical series has shown reaching an all time highof 19.7% in January 2012 and annual average rate of 16.
1% in 2012. According toIndex of economic freedom (2017), Tanzania has made movement towards achievingincome growth and poverty reduction over the past decades. Although, small insize the financial sector has undertaken modernization and credit isincreasingly allocated at market rates, hold up the development of a morevibrant entrepreneurial sector. Regardless of these recent gains, however, theTanzania government appears to lack strong commitment to further institutionalreforms that are important to long term economic development. Long standingformational problems including problems including poor management of public andnot fully developed legal frame work that interferes with regulatoryefficiency. Policy makers in Tanzania might not have been able to hold oninflation at required rates because of their incapacity to regulate thepredictors of inflation and its nature. In this way, any policy authorizationadministered as a cure would be unsuccessful once a wrong diagnosis of theproblem has been made. Therefore, before any measures are taken to cureinflation, it is important that, policy makers take a proper diagnosticapproach to determine the variables that a long-running stationary staterelationship with inflation.
These could include variables like, money supply,interest rate, exchange rate and Growth Domestic Product (GDP). Inflation is aproblem because it lowers incomes, depress saving, makes productive inputs moreexpensive and may act as deterrent to hard work, by that means leading tosub-optimal per person real output growth or economic development (Kyereme2004). Even if there is substantial amount of research of inflation inTanzania, less awareness has been given to forecast inflation by comparingdifferent techniques. Therefore, there is a need for reliable estimate ofinflation in some period ahead. There are many time series techniques that canbe used in forecasting inflation. Each one has its own characteristics,benefits and obstacles. A blog of business forecaster has described thatBox-Jenkins and exponential smoothing models are similar in such a way thatthey are adaptive, can model trends and seasonal patterns and can beautomated.
Autoregressive IntegratedMoving Average (ARIMA), AR (p) is an autoregressive of order p and MA (q) isthe moving average of order q by combining them together they form ARIMA (p,q)which is for stationary series. For non stationary series ARMA becomes ARIMA(p,d,q) with d denoting the number of times the series differenced until it isstationary, they are popularly known as Box-Jenkins model. Exponentialsmoothing attempts to combine the inclusion of all past data by giving greaterweight to more recent observations, R. G. Brown (1962) inverted this technique andit is sometimes called Browns exponential smoothing procedure. In this studyBox-Jenkins and exponential smoothing will be applied to forecast inflation inTanzania.