00:01
Okay, so for a, we're going to construct a correlation matrix, and here is the data that we have.
00:07
And to construct a correlation matrix, we can use the pandas function correlation in python.
00:15
And here is the output that we have, which is the correlation matrix.
00:21
Now for b, we're going to do a multiple regression analysis.
00:28
Now, here is the python code to do the analysis.
00:32
And here we have the output.
00:37
So to find the regression line, we are going to look at the coefficient column.
00:45
So the regression line that we have is now, we want to do the test on the null.
01:33
Βi equals 0 for all i, and alternative of at least 1 is non -zero.
01:46
Now to do this, we want to look at either the f statistic or the probability of f statistic.
01:54
Statistic.
01:54
Now, the p -value that we have is 6 .10 times 10 to negative 5, and we're going to use the significance level of 0 .05, which is larger than the p -value, so we reject the new.
02:19
Now, to see which explanatory variable have slope coefficients that are not significantly different from 0, we can look at the p -value column for x1 through x4, and again we're going to use the significance level of 0 .05.
02:39
Now, from the p -value that we have, we fail to reject the new for x1 and x3, so they are not significantly different from 0.
03:02
Now for c, we're going to remove the explanatory variable with the highest p -value from the model, and here is the code to do it...