A Report On Engle Granger Cointegration Test

The MERI assesses the movements of the Mauritian rupee against the currencies of its major trading partners. There are two indices of the MERI, namely, the MERI1 and the MERI2. The MERI1 is based on the currency distribution of merchandise trade whereas the MERI2 is based on the currency distribution of merchandise trade as well as on tourism earnings. An increase in the index represents a depreciation of the Mauritian rupee whereas a decrease in the index represents an appreciation of the Mauritian rupee. The MERI2 is used as the dependent variable in the regression

To avoid spurious results, unit root tests using Augmented Dickey-Fuller (ADF) (Dickey and Fuller, 1981) and Philips-Perron (PP) are performed to determine the time-series properties of the variables employed in the analysis. Two or more variables are said to be co-integrated when they exhibit long run equilibrium (relationship) if they share common trend(s).Therefore Auto-regressive distributed lag bounds approach (ARDL) is used to test it. The choice is based on several

oux et al., 1987). Finally, we then use these generalized residuals as generated regressors in the

Also, this estimation allows the using of a lagged dependent variable to check whether there is a possible cointegration in the economic variables, with the possibility of one variable forecasting another (Campbell & Shiller, 1988). Including the lagged dependent variable can reduce the occurrence of autocorrelation arising

In this project we first checked consistency and seasonality of S&P500 index stock performance by splitting its recent twenty years historical data into ten two year data and built ARIMA and GARCH models for each sub-period. We found that the models are considerably consistent before 2007-2008 sub-period, and there exists some minor seasonality in several subperiods, but no particular pattern can be identified for the whole period. We then tried to predict future return, volatility and VaR using the model we built for the last sub-period based on rolling forecast procedure. Though the fitted values of 10th sub-period model are

Additionally, I will also be testing for granger causality. In order to test for granger causality though we need to use variables which are stationary.

Considering the following regression model: BRi=β0++β1(Y)+β2(Z)+ui which connects the bank rate (BR) of Canada to foreign exchange rates(Y) and CPI(Z). In this model X1 and X2 are the corresponding independent variables exchange rates and CPI measured in decimals. There were three estimation methods that were used to estimate the model: The Durbin Watson test is used to test the presence of autocorrelation. The residual values from the regression analysis helps determine if there is a relationship between values that are lagged. The result of the Durbin Watson test lies between 0 and 4 and depending on the value it will show the presence or absence of autocorrelation. The value that is closer to 0 indicates that there is positive autocorrelation, 2 indicates that there is no autocorrelation and values approaching 4 indicate that there is negative autocorrelation. For the hypothesis testing I’ve used the F-Statistic testing, in the later section of the paper I will explain my findings and the results.

You need to make use of the inflation formula for the GDP deflator here and compare results between the two years.

Since the great depression, the global economy has been facing a number of ups and downs. With the markets being built rapidly and crashing down in the other instant, elements that actually make up the skeleton of the international economy seems to have correlated variables. These variables seems to have been correlated with each other in numerous combinations and their correlation is what makes the market fluctuate. The volatility of the global market becomes more apparent in times of recession when the correlation between variables like stocks, bonds, US dollar, Euro, gold and oil becomes more apparent. Therefore, it is of paramount importance that these correlated variables should be studied with a very keen eye. Since the link between currencies, commodities, stocks an bonds runs very deep, change in any one of them seems to have a profound impact on the rest hence changing the face of the market.

real GDP and the price level in the short run. If a negative demand shock occurs, then the AD curve will

The DCF cannot account for changes in the current economic system with substantial accuracy. A stable market provides stable betas, but the global economy in October of 1990 had become unstable. Future economic recovery or growth cannot be assumed, nor can continued decline. By analyzing multiple WACCs and IRRs, we attempted to view the case from a

The R-squared of this new model is around 60%, even a little lower than the original model. The statistical significance of other independent variables keeps almost unaffected. The analysis of coefficients in this model with logarithms is not intuitive. So we still stick to the reduced linear model estimated in (a). We analyzed the OLS residuals to check for possible heteroskedasticity for a cross-sectional data set. We computed the fitted values of Price and OLS residuals in our model. Then residuals are sorted by fitted values of Price, and a scatter diagram of residuals is plotted against the fitted values in Figure 2. It is observed that all points are fairly scattered and there is no obvious pattern for the dependence of the residuals on the fitted values. We conclude that a linear model with homoskedastic errors is satisfactory for this data set. We evaluated the validity of this linear regression model by checking the accuracy of its prediction.

I used 16 numbers of lags because we are using quarterly information instead of seasonal. As you can tell by the above graph there is no longer any seasonal data. However, there are 4 or more blue measure points above the red line which indicates there is a time trend. The red line symbolizes Bartlett’s test, which states that a consecutive string of 4 or more spikes above the red line indicates a time trend. I will now take a first difference of the fourth difference of revenue. This is basically taking out both the quarterly and time trend that could be potentially distorting the data. Now on the graph on the right, is the result of taking the first difference of the fourth difference of revenue. It is now very clear that the time trend is no longer existing within the data set, and the quarterly trend has also been removed. It is somewhat concerning that there are spikes at various points. More specifically at the first and fourth lag which I will consider when adding the partial autocorrelation tool. The partial autocorrelation and autocorrelation functions, are used to determine if there is an autoregressive, moving average, or mixed model as I mentioned earlier. The AR represents trend and quarterly values, while MA tends to represent the

The research data was analyzed using a statistical software called STATA. Before running the data for the regression analysis, panel unit-root test was performed to see if data is stationary over time so that it can be used in estimating the variables in question. The value of the variables in the model throughout the research period is not constant; that means, there are some periods where there are spikes. These periods of ups and downs in value of economic variables are called shock in economic jargon. The notion behind testing for stationarity is to identify whether the effect of such shocks is permanent or transitory. If the effect of such shocks is temporary, the subsequent values of the variables will return to their long term equilibrium suggesting that the data is stable even with the presence of shocks.

There has been a long standing controversy among the economist about the validity of PPP (Purchasing Power Parity) in the long run. The parity reveals that prices in two different economies should be identical to each other when they expressed in terms of the same currency. It is a central building block in the monetary models of exchange rate determination. One of the most common practices, to test the validity of PPP is through unit root test of real exchange rate. In this paper unit root test has been done based on the data on Bangladesh and its major trading partner India, to see whether exchange rate has unit root or not. It has been found out that the PPP holds i.e. real exchange is not trend stationary in the