Gas-oil is one of the most important energy carriers and the changes in its prices could have significant effects in economic decisions. The price of this carrier should not be more than 90 percent of F.O.B price of Persian Gulf, legislated in subsidizes regulation law in Iran. Time series models have been used to forecast various phenomena in many fields. In this paper we fit time series models to forecast the weekly gas-oil prices using ARIMA and ARFIMA models and make predictions of each category. Data used in this paperstarted with the first week of the year 2009 until the first week of 2012 for fitting the model and the second week of 2012 until 13th week of 2012 for predicting the values, are extracted from the OPEC website. Our results indicate that the ARFIMA (0.0.-19, 1) model appear to be the better model than ARIMA (1, 1, 0) and the error criterions RMSE, MSE and MAPE for the forecasted amounts is given after the predictions, respectively.

This study is devoted to test the inflation persistence in Iran. For this purpose, respect to the time series data on inflation in Iran (1972-2011), Autoregressive Fractionally Integrated Moving Average model is used. The results of this study show that based on methods of maximum likelihood and modified maximum likelihood degrees of differencing, respectively, are d1=0.482 and d2=0.483. Therefore, based on these findings, the inflation persistence hypothesis is not rejected in Iran. Gradual vanishing of inflation shocks, possibility of inflation is structural and regard to monetary discipline is the most important recommendations of this study.

During the last decades long-memory processes have evolved as an important part of time series analysis and long memory parameter in asset returns has important evidences for many paradigms in modern finance theory. If asset returns display long memory, the series realizations are not independent over time, realizations from the remote past can help forecast future returns. Therefore the presence of long memory in asset returns contradicts the weak form of the market efficiency hypothesis. Also this characteristic has fundamental effect on time series prediction methods. This research examines the presence of long memory in series of return and volatility of Tehran Stock Exchange. Significance evidence of long memory is found in first and second moments of Tehran Stock Exchange return series. Also predictive accuracy of AMRA, GARCH, ARFIMA and FIGARCH models compared in variety of forecast horizons with recursive and rolling estimation schemes. The results of this research show that ARMA model perform better in 1-step ahead forecast, while for greater forecast horizons, including weekly, monthly, seasonal and yearly predictions, FIGARCH model outperform other alternatives.

In this paper we investigate the long memory of Tehran Securities Price Index and fit ARFIMA model using 970 daily data since 1382/1/6 until 1386/4/17. Furthermore, we compare the forecasting performance of ARFIMA and ARIMA models. The results show that the series is a long memory one and therefore it can become stationary by fractional differencing. We obtaine the fractional differencing parameter d=0.4767.Having done the fractional differencing and determination of the number of lags of autoregressive and moving average components, the model is specified as ARFlMA (2, 0.4767, 18). We estimate the parameters of the model using 900 in-sample data and use this estimates for forecasting 70 out-sample data. Comparing forecasting performance of two models illustrate that forecasting performance of ARFIMA model is better than ARIMA model.

 Analyzing time series is a usual way of forecasting various prices.Some of the Analyzing methods are Autoregressive Integrated Moving Average (ARIMA) and Autoregressive Integrated Fractionally Moving Average (ARFIMA). This paper used ARIMA for forecasting weekly gasoline prices. Also the behavior of ARIMA and ARFIMA models has been compared. So for fitting the models we used Stata12 and time series data of Persian Gulf F.O.B. from first of the year 2009 until 26th week of the year 2012. These data are extracted weekly from OPEC website. The results showed ARFIMA (6, .22, 6) has less error in comparison to ARIMA (1, 1, 0), so ARFIMA is better for forecasting the prices.

The study of the effect of memory in different economic indices, especially inflation and money market, has high research attractiveness. In this paper, by using the data of consumer price index for Iran during 1990/04– 2011/11, we investigate the characteristics of CPI’s long–run memory and regress its ARFIMA model. In addition, the amount of error terms in ARFIMA model are examined by FIGARCH model in order to determine what model the heteroscedasticity in inflation is following. The results indicate that monthly time series of inflation may have non-integer root. In other words, the degree of integration for inflation can be a non-integer number rather than an integer. To determine this, an Augmented Dikey-Fuller test, Philips-Prone test and KPSS are used and the results show that the degree of integration for inflation series should lie between zero and one. Thus, the hypothesis of inflation series with memory is proposed. By estimating the parameter of long run memory in the model it becomes evident that the inflation series has the degree of integration of 0.46 and one time differentiating leads to over-differentiation. Hence, inflation series has a long run memory in Iran and the effects of each shock on this variable exists for long periods.