Abstract: -
The concept of joint integration of important concepts in macroeconomic application, the idea of ​​cointegration is due to the Granger (1981), and he explained it in detail in Granger and Engle in Econometrica (1987). The introduction of the joint analysis of integration in econometrics in the mid-eighties of the last century, is one of the most important developments in the experimental method for modeling, and the advantage is simply the account and use it only needs to familiarize them selves with ordinary least squares.

Cointegration seen relations equilibrium time series in the long run, even if it contained all the sequences on the direction of random (non-static, and variables integrated of degree one), they although no sleep will move close together over time and be the difference, including a finger (1), thus the idea of ​​joint integration simulates a balance in the long run to interpret the economic system, is concentrated the main reason for the lack of balance in the inability of economic agents to adapt to economic agents with the information on the fly (2). There are an estimated cost of adaptation and that lead to determining the value of the dependent variable only through the value of current and past some of the independent variables in addition to the evolution of the dependent variable as a result of its interaction with the values ​​of the independent variables current and previous as well as the delay of the variable's in the form of short-term (dynamic). The addition of dependent variables with a delay to the form simplification of the model motor (which tend otherwise to the existence of a strong correlation between the independent variables with a delay) when the restriction (adaptation) values ​​of the dependent variable the current values ​​of the independent variables of delay works to reduce these limits in return for the cost of the border with delay.

On this basis, the application of the methodology, the researcher Engle - Granger (Engle - Granger) in the joint integration of the CT data models for large industrial plants in Iraq for the period (1990-2005) was cross-

sections so that the sector (public, private)

Where the research contained in the theoretical side Mbgesan, in the first part, has been the research methodology by showing the importance of research, which confirmed the account the new addition to the specialists and researchers in this field. The research problem can be summarized in the impact of the application of this method compared to compact way of least squares for fixed effects for the periods and groups, as well as the goal of the research and its premises and the nature of the variables used and where they overlap.

In the second section, the researcher introduced the concept of stability, and how to test the method of combining CT data and time series.
In the practical side, were presented results of the assessment of the variables used in the research and for the period (1990-2005), disaggregated by estimation method and the type of the function of each sector (public, private) separately.

Finally Showing researcher conclusions and recommendations reached by the research.