Regression Analysis

Test ( H 0 , H A , Tc , Ttable , p − value, Stat Conclusion ) whether there is a correlation between

these two variables.

Pearson correlation of Oil index and Spider = -0.124

P-Value = 0.209

HO : ρ = 0 ; There is NO correlation between the oil prices and Spider stock index

HA : ρ ≠ 0; There is a correlation between the oil prices and Spider stock index

The test statistic TC

TC = r x √(n-2)/(1-r2)

TC = -0.124 x √(105 - 2) / (1 – (-0.1242))

TC = -1.268

The test statistics is -1.268

Ttable = ± Tα/2, n-2 = ± T0.025, 103 = ± 1.960

The test statistic Tc does not fall in the rejection region

Also p =0.209 > α=0.05 hence

We do not reject HO and conclude that there is not enough statistical evidence to show

that a correlation exists between the oil prices and Spider stock index.

How do you interpret the sample correlation coefficient?

The sample correlation coefficient cannot be interpreted since there is not enough

statistical evidence to show that there is no correlation between oil index and the Spider

stock index

Is it possible to calculate a correlation coefficient for a sample of teenagers that

measures the correlation between hours of TV watching during the day and whether

or not they completed their homework that day? Explain.

It is not possible to calculate the correlation coefficient between the above variables.

Correlation coefficients are calculated for quantitative or interval variables. Hours of TV

is a quantitative variable and whether or not a student completed their homework is a

nominal or qualitative variable.