AN ECONOMIC STUDY FOR IRAQ ' S RICE IMPORTS FOR THE PERIOD ( 1990-2015 ) and PREDICTION FOR THE PERIED ( 2016 – 2023 )

This research aims at predicting the imported quantities of rice in Iraq for the period 20162023 through the use of the self-regression model (VAR). The values of explanatory variables (local production, population, national income, local price index, of rice consumption), where the predicted values for each of these variables are estimated separately using the moving average method based on the data of the last 10 years. The research concluded that there is no difference between the actual values and the predicted values in the short term. This is the problem that emerged from the research which is the mismatch between these values. Therefore, the research recommended that the short term forecasts be adopted in the formulation of import policies, especially after the validity of the model To predict after testing the predictive power of the model.


INTRODUCTION
The prediction is great importance in the various sciences and fields and plays a great role in drawing the image of the unknown future and trying to plan it based on the various methods of prediction.It is possible to know the importance of prediction by achieving efficiency and effectiveness in the quantities that Iraq needs to import from rice and to know the needs of the country in the short term and to reduce the risks facing Iraq as well as to give a picture of the future direction and contribute significantly to the decision-making and anticipation of its future effects and predictive as a basis for strategic planning and be more effective in the control process and as a basis for the decision the correct administrative and show the importance of paying it in predicting the administration to look into the future and then taking precautions and road towards achieving the desired long-term goals needs him, and receive predictions lights on the road towards achieving the desired long-term goals commodity such as rice of the necessary commodities at the Iraqi consumer table .The problem of study is un knowing the accurate quantities that Iraq may needs in the future .The aim of this study is to predict the quantities that Iraq needs to import from rice in the coming years within the short term so that a suitable import policy can be drawn for an important commodity such as rice is one of the necessary commodities at the Iraqi consumer table (1).The existence of number of independent variables affecting the imported quantities of rice (local production, population, national income, local price, border price and dummy variable) and other variables may be included in the model.

MATERIALS AND METHODS
The prediction has taken several concepts and can be defined as an estimation of the unknown in relation to future events.The path of the phenomenon is examined in the future, so it is a rational attempt to estimate possible future variables by knowing the behavioral and non-behavioral variables of this phenomenon (3).The concept of scientific predicition scientific prediction is important in any area of human life to find a means to help in making current and future decisions, and that it is not without a field of different sciences to find a way to characterize and analyze natural and abnormal phenomena and future prospects.

Predicting using the VAR model.( Vector auto regressive model)
VAR models offer a very simple way of forecasting and are closer to economic reality because their variables (internal and interpretive) interact with each other and therefore must be introduced into the economic system.Working with VAR components at the same time means that the model is shorter and contains fewer decelerations because information expands to include advance information of variables.However, the VAR model predictions are accurate under certain conditions.It is difficult to identify these conditions to avoid error.The number of parameters to be estimated and the degrees of freedom, and that many of the Lagged of the same variable lead to insignificant estimates, perhaps due to multiple linear correlation (11,15) Types of Predictions using VAR 1. Dynamic prediction: It takes several steps starting from the first period of the forecasted sample.Predetermined values are used for the dependent variables to create an expectation of the current values.This option is only available when the estimated equation contains dynamic components.(10) 2. Static prediction: It is calculated in one step using the real and not expected values of the adopted dependent variables when available.In this study, the error correction model was used as one of the methods of this type of expectation.The researchers will predict the values of the explanatory variables and then replace them with the estimated relationship, thus predicting the expected changes in the dependent variable.The average of the expected error is zero, and this does not necessarily mean that the expectation is equal to the realistic values of the variables, because of the error in both the expectation and reality estimates.The importance of prediction(7) From above we note that prediction has great importance in different sciences and fields and plays a dangerous role In drawing the image of the unknown future and trying to plan it depending on the multiple modes of prediction we can summarize the importance of predicting the following(4) points: 1-To achieve efficiency and effectiveness of the institution in flexibility with the external environment.2 Knowledge of the needs of the institution in the short and medium term.3. Reducing the risks facing the institution.4 Give a picture of the institution for its future direction.5. To contribute significantly to decisionmaking and to anticipate its future impact.6.A basis for strategic planning.7 -The basis of the administrative decision is a link between the establishment and its surroundings.8. Establish more effective rules in the control process.9 -To create interdependence, integration and coordination between parts of the establishment.The importance of the Prediction is that it drives the administration to look to the future, and then takes the needs for it, which makes the momentum of the establishment forward more stable and secure, and receive the forecasts highlights the way the institution to achieve goals, which helps to establish a more effective basis for the control, Predicting that the presence of the enterprise in the long run depends on the existence of a continuous demand for its goods or services.Forecasting is a link between the project and the establishment and the surrounding external conditions.

Standard prediction types
The prediction is divided into four types according to the following criteria: (14) First: Duration 1. Ex-Post Prediction The prediction or prediction of the values of the dependent variable in a period following the period in which the model was estimated.During this period the data are actually available for the phenomenon and these predictions are used to compare the actual data with the predicted data to validate the model 2. Ex-Ante Prediction Which is to predict the value of the future variable on the basis of past and present data and information so that no value of this variable has been achieved.

Second: the degree of certainty
Conditional prediction: It is one of the explanatory variables on which the expectation is based on which is not known but must also be foreseen.If this prediction is achieved, the predictions of the phenomenon are realized.2. Unconditional prediction: Prediction based on confirmed information is available on explanatory variables (6).A review of the types of prediction can not be astrology or fiction but a set of qualitative and quantitative methods through which to assess the future of the case under study on the basis of technical and scientific, and as required time and subject.

Prediction methods
It can be noted that prediction methods are divided as follows: First: Systematic Methods (Quantitative Methods): These methods are based on an explicit rule on all explanatory variables that explain the behavior of the phenomenon.Based on the economic theory, all the variables that are included in the interpretation of the phenomenon are determined in the form of an estimable mathematical model, and the systematic methods are the graphical, statistical and mathematical method to reach predictions which are usually less biased and more accurate compared to the qualitative methods (2) .Systematic (quantitative) methods are divided into two types of prediction models: 1. Causal models: This model in which the behavior of the Y variable is explained to some extent by one or more predicted variables.The most important causal models are econometric models, input and output models, simulation models and nonlinear dynamic models (5).

Non-causal models (time series analysis):
These models are based on the historical values of variables that explain their behavior.These include Auto-regression models (AR), moving averages models, exponential models, BaJ models, general trend projections and VAR models.

Second:
Qualitative (Technological) Prediction Methods: This type of method does not require data in the same manner as quantitative prediction methods.The required inputs depend on the nature of the method used, the intuitive thinking of the researcher and his judgment, and the cumulative experience.These methods often require input from a number of trainees specially trained and knowledgeable about the problem.These methods fall into two general categories: exploratory methods and normative methods.From reservations to qualitative prediction methods, it is difficult to measure the accuracy of predictions generated by these methods.Therefore, they are used primarily to provide observations and assist observations, and to help quantitative prediction methods give a clearer picture rather than providing them with specific numerical predictions (8).

Measure the predictive force of the estimated models
After studying the variables of the import function of the rice crop for the period 1990-2015, we obtained through the results of the error correction vector model to determine the most important factors affecting the rice import function.To know the model's ability to predict, the actual values will be compared with their estimated values by the estimated model using the following table and the following figure respectively.---Fromtable 1. and figure 1. , we can see that the estimated values via using the error correction model for rice import quantities are only far from actual values in the years (1997, 2010, 2011, and 2015), because of increased imports due to low production and water scarcity, Improved living and increased per capita income after 2003 , as well as the fact that rice is one of the most essential and most essential foodstuffs on which the Iraqi individual depends on his daily meals.

Table 1 .
The estimated values of the import quantities of rice in Iraq for the period 1990 -2015 Source: From the researchers work based on the statistical program (Eviews 9).

Figure 1 .
Figure 1.The comparison between the actual value curve and the estimated values of Iraq's rice import guintites for the period 1990-2015 Source: From the researchers work based on the statistical program Eviews 9.

Figure 2 .
Figure 2 .the results of the test of the variance coefficients for the base and the root mean square error of the dynamic model (at the level) for the period 1990-2015 Source: From the researchers work based on the statistical program Eviews 9. while table 3. shows the results of the TIA and the square root error of the dynamic model (at the first difference) for the period 1996-2015

Figure 4 .
Figure 4. the temporal evolution of the actual values and predictive values of the imported quantities of rice for Iraq for the period (1990-2023).Source: The researchers worked according to the statistical program (Eviews 9).We can note from the figure that the estimated values are not far from the actual values during the period 1990 -2023 except the years 1997, 2010, 2011 and 2015 , in which imports were are larger quantities than the other years under study and for the reasons mentioned earlier of the lack of local production and the increase in population and water scarcity After selecting the model that has a high predictive capacity, it will be used to predict