Page 67 - Azerbaijan State University of Economics

Murad Yusifov: Modelling the inflationary processes and forecasting:an application of ARIMA,
SARIMA models

competitive market principles without political and administrative interference in such case the

made forecasting will fit the forecasted data in future. Because without interference there all

processes is randomly developped in the framework of economic system.

1. Research works in the world

There several research work in this field. Ezekiel N.N.Nortey,Benedict Mbeah-Baiden,

Julis B.Dasah and Feliks O.Mettle forecasted the inflation with ARCH modelling[1].

S.O.Adams, A.Awujola, A.İ.Alumgudu used the ARIMA models to forecast the inflation rate in

Nigeria and got the adequate results [2]. Ekpenyong, Emmanuel John, Omekara C.O. forecasted

the inflation using the periodoqram və Fourier series[3]. Sani Doguwa və Sara O.Alade used the

SARIMA model to forecast the inflation in Nigeria[4].

2. Methodology and data

a. Theoretical framework and methodology

There are five different economic forecasting approaches based on times series data.

1. Exponential smoothing methods

2. Single equation regression models

3. Simultaneous equation regression models

4. Autoregressive integrated moving average models(ARIMA)

5. Vector autoregressiion

Forecasting on simultaneous equation regression models has subsided because of their

poor forecasting performance, especially since the 1973 and 1979 oil price shocks because of

OPEC oil embargoes and as well as because of Lucas critique [5,p. 837]. The essence of this

critique is that the parameters obtained from an econometric model are dependent on the policy

prevailing at the time the model was estimated. So, model will change if there is a policy change.

67