GT = c(0.55,0.51, 0.8, 0.74,0.02,0.58,0.47,0.45,0.99,0.13)
Reg1 = c(0.46,0.39,0.87,0.86,0.1,0.68,0.55,0.53,1.09,0.22)
Reg2 = c(0.66,0.41,0.92,0.65,-0.12,0.45,0.59,0.34,0.86,0.22)

#calculate residual
diff1 = GT - Reg1
diff2 = GT - Reg2


#difference between the two residual regressors
diff = diff2 - diff1
diff
##  [1] -0.20 -0.02 -0.05  0.21  0.22  0.23 -0.04  0.19  0.23  0.00
#mean difference
mean_diff = mean(diff)
mean_diff
## [1] 0.077
#standard deviation
std_dev = sqrt(sum((diff-mean_diff)^2) / (length(diff)-1))
std_dev
## [1] 0.1562086

\(\sigma = \sqrt{\frac{1}{n-1} \sum_{i=1}^{n} (diff_i - \mu)^2}\)

#t-value 
tvalue = mean_diff/(std_dev/sqrt(10))
tvalue
## [1] 1.558784

\(t = \frac{\text{Mean difference}}{\frac{\sigma(Diff)}{\sqrt{n}}}\)

#t-table significance level at 5% 9df is 2.262 
sig_value = 2.262

#is it significant?
tvalue> sig_value
## [1] FALSE

From our calculations, we observe the t-value is positive. This implies the first term which is regressor 2 has a higher mean compared to regressor 1. However this t-value is not significant enough at the 5% significance level to conclude definite difference.