---

Rice blast, caused by Pyricularia grisea, is one of the most important diseases of this crop in Iran and all over the world. To evaluate the relationship between spore population (SP) and meteorological factors, SP was measured daily using spore trap during growing seasons of 2006-2008 in Rasht and Lahijan regions (Guilan province, Iran). Weather data including precipitation, daily maximum and minimum temperatures, daily maximum and minimum relative humidity and duration of sunny hours were obtained from weather stations which were five kilometers away from the fields. The relationship between spore population and metrological factors was evaluated by Neurosolution 5.0 software. Weather data and spore population were considered as input and output data, respectively. In this study, multilayer perceptron neural network, regression model and Log(x + 1) transformation were performed. To evaluate the model efficiency, correlation coefficient and mean square error were used. The results showed that the correlation coefficient (r) and mean square error (MSE) parameters were 0.55 and 0.03 in Rasht and 0.1 and 0.03 in Lahijan, respectively. The results also showed the potential of this model for modeling SP using meteorological factors; however more data is needed for validation of this model. There has been no previous report on modeling the relationship between SP and meteorological data using artificial neural network in Guilan province (Iran).

---

Applying a precise forecasting method is necessary to achieve acceptable results in IPM programs. Performances of the wing and delta pheromone traps for forecasting the codling moth phenology were compared with physiological time data based on Degree-Hours. Six pheromone traps (three wing and three delta style) were applied for the monitoring of the codling moth population. Traps were placed in an apple orchard in Tehran Province, Damavand region by the start of bloom. All traps were checked every week and the number of moths caught was recorded. Physiological time was estimated by using hourly recoded temperature, considering temperature thresholds for codling moth development. The results showed that the delta style traps statistically caught more male moth than wing traps. It was also shown that the results of the pheromone traps data were affected severely by weather conditions. Moreover, false fluctuations in recorded data from pheromone traps made some false population peaks, the interpretation of which was very hard. On the other hand, forecasting model based on the physiological time data, was not affected by the mentioned conditions and its results was easy to use for determination of the pest phenology without further interpretations.  

---

Lack of a structured anticipation about different aspects of high usage product of the national petrochemical company, has forced this company to buy published anticipated prices from foreign countries. Prevent the outflow of foreign exchange and tolerance of political factors, such as sanctions in this field, require a prediction of prices in Iran. Due to chain-like nature of petrochemical products and the absence of precise knowledge of effects of many factors on price, researchers are forced to solve problems with high complexity and high grade of equations. Selecting number and type of input variables of neural network has a significant impact on the performance of a system. Therefore fundamental analysis relying on theory of supply / demand and macroeconomic perspective alongside of Delphi statistical method were used to select the most influential factor. This factor is the price of petroleum products. At First, the overall topology of the neural network is designed using controlled variables, then, considering the independent variables, the optimal network has selected. After creating the user interface, communication of system with optimal neural network was established. To evaluate the actual price of considered product in reference year, it compared with the prices predicted by the proposed system and purchased prices predicted from CMAI; acquired results proved acceptable effectiveness of the proposed system with less than 3% error in predicting of considered chain. Using this system can result in petrochemical companies’ independency from buying forecasted prices from foreign companies and prevent exiting currency from country.

---

The present study aims at applying different methods for predicting spring inflow to the Amir Kabir reservoir in the Karaj river watershed, located to the northwest of Te-hran (Iran). Three different methods, artificial neural network (ANN), ARIMA time se-ries and regression analysis between some hydroclimatological data and inflow, were used to predict the spring inflow. The spring inflow accounts for almost 60 percent of annual inflow to the reservoir. Twenty five years of observed data were used to train or calibrate the models and five years were applied for testing. The performances of models were compared and the ANN model was found to model the flows better. Thus, ANN can be an effective tool for reservoir inflow forecasting in the Amir Kabir reservoir using snowmelt equivalent data.

---

The evolution of financial data shows a high degree of volatility of the series, coupled with increasing difficulties of forecasting financial variables. Some alternative forecasting methods, based on the literature review, have been developed, which can be particularly useful in the analysis of financial time series. Despite of the numerous time series forecasting models, the accuracy of time series forecasting is fundamental to many decision processes. Selecting an efficient technique in unique situations is very difficult task for forecasters. Many researchers have integrated linear and nonlinear methods in order to yield more accurate results. In practice, it is difficult to determine the time series under study are generated from a linear or nonlinear underlying process while many aspects of economic behavior may not be pure linear or nonlinear. Although both ARIMA and Artificial Neural Networks (ANNs) models have the flexibility in modeling a variety of problems, none of which is universally the best model used indiscriminately in every forecasting situation. In this paper, based on the foundations of ARIMA and ANNs models, a hybrid method is proposed to forecast exchange rate. Empirical results indicate that integrating linear and nonlinear ARIMA and Artificial Neural Networks (ANNs) models can be an effective way to improve forecasting accuracy achieved by either of the above linear and nonlinear models used separately.

---

---

The inflorescence rot is an essentially high impact (or damaging) disease of date palm. The current research was carried out to help develop a decision-making system in Abadan, Khorramshahr, Shadegan, Ahwaz, Mahshar, and Behbehan regions of Khuzestan province Iran based on climatic and geostatistical models using five-year data from 2011 to 2015. Samples were taken randomly from 10 date palm trees within one orchard in each of 33 villages. The disease started in March, and the damage reached its peak values in April. The forecasting model of damage factors has been significant at levels 1 and 5%. The model nuggets for disease in Abadan-Khorramshahr, Shadegan, Ahwaz, Mahshar, and Behbehan regions were 2.1, 1.1, 0.09, 2.60, and 0.27 km, respectively. These results show that the disease damage estimation errors were low at distances less than within sampling space. The effective ranges of variograms were 4.9. 8.3, 9.1, 5.1, and 4.2, respectively, indicating the disease distribution in the region. The sill of models were 0.41, 0.46, 0.46, 0.29, and 0.58, respectively, indicating that correlations between the damage data were at the lowest level and could be monitored at distances more than these thresholds. Findings are fundamental steps in creating a decision-making system in the date palm protection network. Therefore, it could be concluded that the date inflorescence rot disease can be monitored, forecasted, and controlled correctly before the maximum damage occurs.

---

The forecasting of hydrological variables, such as streamflow, plays an important role in water resource planning and management. Recently, the development of stochastic models is regarded as a major step for this purpose. Streamflow forecasting using the ARIMA model can be conducted when unknown parameters are estimated correctly because parameter estimation is one of the crucial steps in modeling process. The main objective of this research is to explore the performance of parameter estimation methods in the ARIMA model. In this study, four parameter estimation methods have been used: (i) autocorrelation function based on model parameters; (ii) conditional likelihood; (iii) unconditional likelihood; and (iv) genetic algorithm. Streamflow data of Ouromieh River basin situated in Northwest Iran has been selected as a case study for this research. The results of these four parameter estimation methods have been compared using RMSE, RME, SE, MAE and minimizing the sum squares of error. This research indicates that the genetic algorithm and unconditional likelihood methods are, respectively, more appropriate in comparison with other methods but, due to the complexity of the model, genetic algorithm has high convergence to a global optimum.

---

A critical step to develop artificial neural networks that has considerable effect on network performance is designing architecture of neural networks. In designing the architecture of networks, generally, the number of hidden layers, number of neurons in each layer and transfer functions are determined. Most researchers often use trial and error approach and/or ignore interactive effects between the factors of design. In this research, a model is presented based on the design of experiment (DOE) for optimal architecture of neural networks. The proposed model was applied to determine the optimal architecture of neural network for forecasting the monthly consumption of gas oil of Iran. To evaluate the effectiveness of the proposed model, using the common method of trial and error was used and advantages of the proposed model were shown. In addition, to compare the performance of neural networks by statistical methods, two models based on regression and ARIMA were designed. Comparison of the forecasting results obtained by neural networks and the statistical methods proved that the proposed model produced better forecasts in all performance criteria.

---

In this paper, the energy demand of transport sector from 1386 to 1400 was forecasted using artificial neural networks (ANN) approach considering economic and social indicators. Feed forward supervised neural networks to forecast and back propagation algorithm to train networks were used. In order to analyze the influence of economic and social indicators on energy demand of transport sector, Gross Domestic Product (GDP), population and the total number of vehicles in 1347-1385 were taken into consideration. The obtained results as compared with the multiple regression method, revealed much less mistakes. The average absolute error percentage was decreased from 15.52% to 6.05%. .

---

National accounts data are of the most important statistical tools in planning and making economic policy. Therefore, forecasting the main economic variables in the economy is of great importance. Economic growth is one of the key macroeconomic variables, which gets top priority in forecasting.  The purpose of this study is to identify the appropriate methodology for forecasting economic growth in Iran. This study introduces fuzzy regression model and its’ ability to forecast economic growth of Iran in comparison with Error Correction Model (ECM). To do this, the Iran’s GDP is modeled through ECM and Fuzzy regression models using annual data form 1959 to 2001. Then, Iran’s GDP growth is predicted for 2002-2012. Finally, the performances of these models are compared using common criteria for evaluating forecast accuracy including mean absolute error (MAE), root mean square Error (RMSE), mean absolute percentage error (MAPE) and Theil’s inequality coefficient (TIC). The results indicate that the performance of fuzzy regression is far better than that of ECM in predicting GDP growth in Iran. Moreover, forecast accuracy of fuzzy regression model is of statistically significant difference in comparison with ECM model.

---

Abstract This paper presents a neuro-based approach for annual gasoline demand forecast in Iran by taking into account several socio-economic indicators. To analyze the influence of economic and social indicators on the gasoline demand, gross domestic product (GDP), population and the total number of vehicles are selected. This approach is structured as a hierarchical artificial neural network (ANN) based on supervised multi-layer perceptron (MLP), trained with back-propagation (BP) algorithm. This hierarchical ANN is designed properly. The input variables are GDP, population, total number of vehicles and the gasoline demand in the last one year. The output variable is the gasoline demand. The paper proposes a hierarchical network by which the inputs to the ending level are obtained as outputs of the starting levels. Actual Iranian data between 1967 and 2008 were used to test the hierarchical ANN hence; it illustrated the capability of the approach. Comparison of the model predictions with validation data shows validity of the model. Furthermore, the demand for the period between 2011 and 2030 is estimated. It is noticeable that if there will not be any price shock or efficiency improvement in the transportation sector, the gasoline consumption may achieve a threatening level of about 54 billion liters by 2030 in Iran.

---

The overall objective of the present paper is demonstrating the utility of price forecasting of farm prices and validating the same for major crops namely, Paddy, Ragi and Maize in Karnataka state for the year 2016 using the time series data from 2002 to 2016. The results were obtained from the application of univariate ARIMA techniques to produce price forecasts for cereal and precision of the forecasts were evaluated using the standard criteria of MSE, MAPE and Theils U coefficient criteria. The results of ARIMA price forecasts amply demonstrated the power of the ARIMA model as a tool for price forecasting as revealed by pragmatic models of forecasted prices for 2020. The values of MSE, MAPE and Theils U were relatively lower, indicating validity of the forecasted prices of the three crops.

---

The aim of this paper is to develop a prediction model of energy demand of Iran’s industrial sector. For that matter a Markov Chain Grey Model (MCGM) has been proposed to forecast such energy demand. To find the effectiveness of the proposed model, it is then compared with Grey Model (GM) and regression model. The comparison reveals that the MCGM model has higher precision than those of the GM and the regression. The MCGM is then used to forecast the annual energy demand of industrial sector in Iran up to the year 2020. The results provide scientific basis for the planned development of the energy supply of industrial sector in Iran.

---

In recent decades, due to the importance of future values of macroeconomic variables, a range of predicting methods and models has been studied and evaluated. The main purpose of this paper is to compare different methods of predicting Iran's economic growth using seasonal time series data during 1990-2017. To this end, economic growth is predicted using dynamic model averaging (DMA), dynamic model selection (DMS), BMA, BVAR, TVP and AR models in three prediction horizons (one, four and eight seasons). The models used in this study are categorized into three spectra, large-scale (including 112 variables in nine factor blocks), average-scale (including 10 variables) and univariate models. The results show that the predictions of DMS and DMA are more efficient than other traditional prediction. 

---

In general, energy prices, such as those of crude oil, are affected by deterministic events such as seasonal changes as well as non-deterministic events such as geopolitical events. It is the non-deterministic events which cause the prices to vary randomly and makes price prediction a difficult task. One could argue that these random changes act like noise which effects the deterministic variations in prices. In this paper, we employ the wavelet transform as a tool for smoothing and minimizing the noise presented in crude oil prices, and then investigate the effect of wavelet smoothing on oil price forecasting while using the GMDH neural network as the forecasting model. Furthermore, the Generalized Auto-Regressive Conditional Hetroscedasticity model is used for capturing time varying variance of crude oil price. In order to evaluate the proposed hybrid model, we employ crude oil spot price of New York and Los Angles markets. Results reveal that the prediction performance improves by more than 40% when the effect of noise is minimized and variance is captured by Auto-Regressive Conditional Hetroscedasticity model.

---

Dependency of Iranian Economy on oil revenues has provided conditions for imposing further sanctions on Iran. One way for Iran to get rid of sanctions is to sell its oil in currencies other than US dollar. In this regard, this article evaluates the risks for Iran if it, in selling oil, substitutes US dollar with currencies of its oil importing countries. We firstly apply Autoregressive Integrated Moving Average (ARIMA) and Self-Exciting Threshold Autoregressive (SETAR) models on Yuan and Rupee data for the period of 1990:01-2019:05 as well as on Euro data for the period of 1999:01-2019:05 and then based on the estimated models, forecast losses and gains for the period of 2019:06-2021:12 if Iran sells oil to China, India and Europe and receive payments respectively in Yuan, Rupee and Euro. Our forecasts indicate that selling oil to India and China and receiving oil revenue in Rupee and Yuan respectively will significantly decrease value of oil exports in range of 5-23 percent due to very likely devaluation of these currencies vs. the US dollar. Therefore, Iran must firstly use in its oil transaction relevant diplomacy with its oil importing countries, requesting them to share in risks of devaluation of their currencies vs. US dollar. Secondly, as a particular example, this article shows that political decisions may bring in economic consequences for the country. Therefore, Iranian authorities are expected to consider economic consequences of their political decisions more seriously and with sufficient transparency.

---

Saffron is one of the most valuable agricultural and medicinal plants of the world and has a special place in Iran's export of products. Presently, Iran is the world's largest producer and exporter of saffron and more than 93/7% of the world production belongs to Iran. However, despite the long history of saffron cultivation and its value-added in comparison to many of the other crops in the country, a lower share of new technologies is assigned to it, and its production is mainly based on local knowledge. This study aimed to develop and evaluate the performance of Adaptive Neuro-Fuzzy Inference System model (ANFIS) in calculating the yield of saffron using meteorological data from 20 synoptic stations in the province, including evapotranspiration, temperature (maximum, minimum), the mean relative humidity, and rainfall. To this end, by using software Wingamma, data and parameters were analyzed and the best combinations of inputs to the model were determined. In order to assess the models, statistical parameters of correlation coefficient, the mean absolute error, and mean square error were used to predict the performance of the plant. ANFIS model was most effective when the data of total minimum temperature, precipitation, evapotranspiration, and relative humidity of autumn were used as independent variables for forecasting yield (R²= 0.5627, RMSE= 2.051 kg ha^-1, and MAE = 1.7274 kg ha^-1) .

---