Evaluation of the Main Macroeconomic Indicators and Their Role in the Economic Growth of Albania

This project it is focused in the link between some macroeconomic indicators as an important factor of sustainable growth and the development economy. Specifically in this study is given a presentation of inflation on years in our country, giving the factors that have influenced in it. Also we have received some data from government statistics: Dependence of the nominal stated interest rate on loans from , which used to measurement of (GDP) Gross Domestic Product. Potential product measured using the linear regression method. The results of the methods were compared using time series analysis, for measurement of economy cycles and their intensity. Economic stability is a necessary condition for sustained economic growth of a country and for improving its welfare in the long term. It is also evidenced by the experience of different countries. As the experience as well as theory have shown that the establishment and maintenance of equilibrium within and between sectors of the economy is a necessary condition of economic growth.


Introduction
Specifically this study gives a presentation on inflation rates by years in Albania and the factors that influenced it.Also, data from government statistics such as dependence of the nominal stated interest rate on loans from the consumer price index are used to measure inflation and measurement of (GDP) Gross Domestic Product.Here they emerge to model the variables which consumers.
The CPI one of indicators which is used to measure inflation.

Gross domestic product
The data are for the period Q1 1995 -T2: 2011 taken by the Ministry of Finance, statistical INSTAT ect..The Econometric program that is used is EViews7 and the variables we have noticed: 1.The nominal interest-rate loans 2. CPI-Price Index consumer, which serves to measure inflation 3. GDP level of Gross Domestic Product The model that we will present below is the Bank of Albania's efforts to modeling one of this.

General Issues
In our study we have analysed the relationship between GDP inflation -interest and their role in the economic growth for the Albania's case.Also this study have analyzed some econometric models, to see the connection that exists between variables, and also to test the statistical significance of the models.

Material and Methods
In this study we have received some data from government statistics: dependence of the nominal stated interest rate on loans from, which used to measurement of (GDP) Gross Domestic Product.E-ISSN 2281-4612 ISSN 2281-3993 Academic Journal of Interdisciplinary Studies MCSER Publishing, Rome-Italy The Econometric program that was used was EViews7 and the variables we have noticed: The data are for the period Q1 1995 -T2: 2011 taken by the Ministry of Finance, Statistical , INSTAT.
Statistical Reports of the Ministry of Finance.Publications by Instad, some topic by "Econometrics" module, some topic by "Money and financial institutions module" ect..

Testing the Model with Durbin-Watson Test
Durbin-Watson test that is more useful test for detecting the serial correlation known as autocorrelation.Durbin-Watson test moves from (0, 4).To perform the test, we build hypotheses: H0: = 0 (no autokorelacion) Ha: 0 (no autokorelacion) Durbin-Watson stat in the value 1.018 testifies to the value of the non-existence of autocorrelation (we know that DW2 = 2 or near 2, the base hypotheses is H0: = 0 that means not has autokorelacion).
Assessment of simple linear relationship between interest rates and gross domestic produktitit From the table of regression we can see that tv = -3.42> tk=1.96H0 In this case we say that the basic hypothesis falls down and our model is statistically significant.since we are in the two variable model to the same conclusion we will come and if we analyze the table ANOVA and Fisher criteria where if Fv> Fk then we say that the basic hypothesis is rejected (H0 ) and accept the alternative hypothesis (Ha ) that model 2281-4612 ISSN 2281-3993 -3993 Academic Journal of Interdisciplinary Studies MCSER Publishing, Rome-Italy is statistically significant.The opposite if Fv <Fk which means that the model is not statistically significant Ho .
In our case we can see that Fv=11.71>Fk=5 which means that Ho and our model is statistically significant.Durbin-Watson stat in the value 1.177 testifies to the value of the non-existence of autocorrelation (we know that DW2 = 2 or near 2, the base hypotheses is H0: = 0 that means not has autokorelacion).

The Multiple Model of Regression
The model in which the number of explanatory variables is greater than regression model.Let The relation that exists between the variables is given by the following equation: I=21.592+0.5738*ICK-0.0094*GDPIf CPI increase by one unit and the others variables independently are constant ,then the interest rate will increase by 0.5738 units If GDP increases by one unit and the others variables independently are constant, then the interest rate will decrease by 0.0094 points.
If all independently variables are 0 then the interest rate is 21.592

Let do the Interpretation of multiple regression model
If in the model we do increase the number of explanatory variables, the coefficient of determination R2 will only increase.
This is because with the increasing number of explanatory variables, the error will be reduced, since the sum of the squares errors (SKG) decreases.The fact that SKT = SKR + SKG, when amount of the squared errors decreases, the amount of squares regression will increase and consequently the coefficient of determination R2 will increase.
In the case of multiple regression model it is necessary to correct the determination coefficient to eliminate the overvaluation and therefore is used the corrected determination coefficient which is obtained by dividing the amount of squares regression with the degrees of freedom .
Then in our case the corrected coefficient of determination is 0.76, which means that 76% of the variance explained by inflation and GDP and 24% of the variance explained of other variables that are not included in the model.
From this higher percent % of this coefficient expect our model to be statistically significant.

We do the test to show how important is model
We will do testing with Fischer, seeing that Fv = 26.91>Fk = 5 means that Ho and our model is statistically significant , based on the highest percent of adjusted determination coefficient.
We do the test of statistical significance of partial coefficients regression.
With the help of Fisher criterion we conclude that whether or not the connection between variables was important.
Namely that at least one of the parameters was different from zero, then it is interesting to carry out a new test to evaluate the parameters that are important, but always knowing that the remains variables have normal distribution .We build hypotheses: 2-Consumer Price Index, independent variable Ho: 2 = 0 isn`t important Ha: 2 0 is important The null hypothesis will be rejected if tv > tk ku where in our case TV = 5.907> tk = 1.96 that brings us to the basic hypothesis is rejected.This means that the partial coefficient of regression 2 Consumer Price Index is statistically significant.
3-Gross Domestic Product, independent variable Ho: 3 = 0 isn`t important Ha: 3 0 is important In our case -2.619 tv => that tk = 1.96,The null hypothesis will be rejected means that 3 the partial coefficient at GDP is statistically significant.
We conclude that: Since all partial regression multiple coefficients are statistically significant , after testing of their statistical significance then we say that our model is better.

Wald Test
Let do the testing of equality of two partial coefficients of model.
We are doing the testing the hypothesis if that 2 and 3 are equal.Prepare the assumptions: Ho: 2= 3 ose 2-3=0 Ha: 2 3 ose 2-3 0 For testing the hypothesis will use student test: In this case we see from the table that TV = 5,057> tk = 1.96, then the null hypothesis is rejected and thus mean that the partial coefficients 2 and 3 are not equal.

Results and Discussions
In this study we have analysed the relationship between both real and nominal interest rate.In our study we have analyzed some econometric models, to see the connection that exists between variables.Also we have analysed the multiple regression.From result our model was statistically significant.
If GDP is 0 then the interest rate is 31,019 The coefficient of determination R2 (R-squared) is calculated by formula And it shows us the percentage of dependence variance variable NPL / TKB explained by all the variables together.But in our case we have ready from the table of regression the value of 0.438.So 44% of the interest rate variance explained by GDP and 56% is explained by other variables not included in the model.
The bond that exists between these two variables is expressed by the equation I = 31.019-0.0175*GDP We now do the simple linear interpretation :If GDP increases by one unit then the interest rate will decrease by 0.0175 points.
's see what happens if we include all independent variables on the model in simultaneously.