I choose to discuss the change in Gross Domestic Product (GDP) from years 1970-2000. GDP is a very important tool in measuring the economic condition of a country. When GDP is increasing from year to year, it would be safe to assume income, labor opportunities, and overall business would increase as well. I decided to discuss the change of GDP opposed to total GDP, for an understanding of the changes in the select macroeconomic variables that significantly affect GDP. The purpose of this study is to attempt to determine what factors affect the change in GDP, and what arrangements can be made to result in more accurate data. This can be helpful to any nation’s economy. If accurate results ...view middle of the document...
This data is time series data from 1970-2000.
These are the results of the Multiple Regression of the three dependent variables to GDP, resulting from data in Appendix B.
From the results above we notice a very strong correlation. The R^2 is significantly high with .9582 of GDP being explained by this equation. At a 5% alpha level, the P-value of .0001 shows that the correlation is significant and that the equation explains the change of GDP to a great extent. For the regression coefficients all variables are significant. The T-score for consumption is quite high, which makes it safe to assume that a 1 million dollar increase in Consumption will result in a 807,287 dollar increase in GDP, (1,000,000(.807287)). The T-score for Investment is also high enough to be a significant variable. A 1 million dollar increase in Investment will result in a 585,340 dollar increase in the rate of GDP, (1,000,000(.585340)). The T-score for Unemployment rate is also found to consider the variable significant to GDP. Assumed as before the relationship is negative, for every 1 % increase in unemployment rate will result in a loss of 401,595.02 dollars in rate of GDP (.01(40.1595)*(1000000). The Durbin Watson for this regression analysis is 2.019, which gives the notion of no autocorrelation. Looking at Appendix C, at the cross correlation matrix, the values are insignificant and small representing no multi-collinearity suggesting that the independent variables have no relation to each other. Appendices D-F represents a linear line model.
I found that the refinement step lead to less accurate data then my first regression run. Referring to Appendix H the T-score for CC (C/GDP) provides for an insignificant relationship. There has not been any effect found for CC on GDP.
The R^2, Ra^2, and the DW values are very close to what was found in the...