Exercise

Hypothesis test for the correlation

Still consider the constant expected return model (CER) that was introduced in exercise 2.

You would like to test for each \(\rho_{ij}\) (\(ij =\) "VBLTX, FMAGX", "VBLTX, SBUX" and "FMAGX, SBUX"):

$$H_0: \rho_{ij} = 0 \text{ vs. } H_1: \rho_{ij} \neq 0,$$ using a 5% significance level. In other words, you would like to investigate whether the correlation between two return series is significantly different from zero according to the data. Perform the test for correlation between paired samples at the 95% confidence level. You can use the R function cor.test() for this problem.

Instructions

100 XP
  • Use the cor.test() function to test whether the correlation between the returns of "VBLTX" and "FMAGX" is significantly different from zero. What do you conclude?