How Gdp Affect Mauritius- an Econometric Expla

Only available on StudyMode
  • Download(s) : 40
  • Published : March 14, 2013
Open Document
Text Preview
Regression Analysis
Variables| Coefficient| Standard Errors| T-Ratio| P-value| In Real interest rate| -0.2600846| 0.1930792| -1.35| 0.196| In Inflation rate| 0.0972735| 0.1828835| 0.53| 0.602|
In Money supply| 0.9009881| 0.897232| 10.04| 0.000|
No constant| -0.0138525| 2.724178| -0.01| 0.996|
Dependent variable: Semdex

Prob > F = 0.0000
R- Squared = 0.8841
Adj R-squared = 0.8636


The Adjusted R2 - the coefficient of determinationR2 is used in the context of statistical models whose main purpose is the prediction of future outcomes on the basis of other related information. The R-squared is 0.8841, when adjusted being about 0.8636 meaning that approximately 86% of the variability of LSEDX is accounted for the variables in the model as the adjusted R-squared attempts to yield a more honest value. As R-squared gets closer to 100%, it means that the model is 100% right. This shows that this model is reliable at 88% and it explains any variation in the dependent variables.

The economy has progressed significantly over the past decades. Indeed, Mauritius has gone up the scale from a least developed country to a developing country status. This has been possible due to a certain extent because of the control of money supply, inflation and real interest rate on savings.

Testing for Multicollinearity

| Lnsemdex| Lnreal Interest Rate| Lninflation| Lnmoney Supply| Lnsemdex| 1.0000| | | |
Lnreal interest rate| -0.1669| 1.0000| | |
Lninflation| -0.3325| -0.3303| 1.0000| |
Lnmoney supply| 0.9289| -0.0303| -0.4483| 1.0000|
Before checking for multicollinearity is important as it helps to know whether there is a strong correlation amongst the explanatory variables in a regression model. The results above show that there is highest correlation between Lnsemdex and Lnmoney supply at 93%. Correlations around 0.8 and 0.9 are not desirable as they will indicate multicollinearity. But in our case, it could be an exception as the highly correlated variable which is LM2 is due to the fact that the money supply has continuously been increasing during 1990 and 2010.

Theoretically, the money supply has a negative impact on stock prices because, as money growth rate increases, the inflation rate is also expected to increase thus leading to a fall in stock price. However, an increase in the money supply would also boost up the economy and hence corporate earnings would increase. This would lead to a rise in future cash flows and stock prices.

Government measure the money supply to gain information about trends in the aggregate demand, the state of financial markets and the need for and effectiveness of monetary policy. M2 increase due to increase in bank lending, government borrowing and a net inflow of money into the country.

Tests for Reliability of Regression Results

Variable| VIF| 1/VIF|
ln Inflation| 1.47| 0.680621|
ln Money supply| 1.31| 0.763267|
ln Real interest rate| 1.17| 0.851065|
Mean VIF| 1.32|

Multicollinearity is the existence of an exact linear relationship among all the explanatory variables in a regression model. There are various reasons explaining why multicollinearity exists. The main reason why multicollinearity exists mostly in time series data is due to the fact that the regressors share a common trend, thus leading to high correlation among them.

To check for multicollinearity the variance inflation factor (VIF) is interpreted. As a rule of thumb, a variable whose VIF values are greater than 10 may merit further investigation when it comes to multicollinearity. All the regressors in our model do not have the signs that accord with prior expectations as our Mean VIF generates a value of 1.32.

Testing For Autocorrelation

In case autocorrelation exists in a model, then it means that it is not in line with the assumption classical linear regression model due...
tracking img