BUSS1020 Written Assignment Essay

1046 Words Mar 24th, 2014 5 Pages
BUSS1020 Written Assignment
Task 1
According to the data downloaded from the website, it involved some information about the all ordinaries index. For instance, it covered the date, open, high, low, close, and volume and adjust close.
In the question, it defined date 1 as the first trading day of the study period. Date 0 which is one day before the first trading day Date 1. First of all, using excel to set up the time index column to arrange the date of the trading day. The range of the data is from date 0 to 1087.
In order to calculate the daily return for the all ordinaries index, the formula could be used. is the return for each day, stands for the all ordinaries index at a specific date t and is the data of the all
…show more content…
Also, the intercept of the regression shows when the time index is 0 (the day before the first trading day), the average price of the all ordinaries is $4191.64.
Moreover, the R square (r2) is called the coefficient of determination. It has a specific region which is. The value of r2 in this situation is computed as 0.131787415 which is involved in the region. Comparing the r2 =0.131787415 with 0 and 1, it is obvious that the r2 value is closer to 0. It indicates that the linear relationship between time index and all ordinaries index is not strong or weak. In addition, the regression model captured less variation.
Furthermore, Durbin-Watson (DW) test is applied to check the independency of the residuals over time. H0 assumes residuals are not correlated and H1 assumes positive autocorrelation exist. By using excel, the result of DW test is 0.013607965. In contrast to the critical value dL= 1.899 and dU = 1.904, DW is less than dL which reject the H0 and claim that the autocorrelation exist. This result illustrates that the residuals are dependent which is not good.
It is also shows through the residual plot below, the dots on the residual plot haven’t formed a straight line. It is said that the errors are not independent.

A satisfied regression assumption is that the two variables have linear relationship.
T test are used to confirm whether the

Related Documents