The residual value is the difference between the observed/given value of the dependent variable (y) and the predicted value (ŷ).
i.e. residual value = given value - predicted value
From the given table, g is the residual value where the given value is 13 and the predicted value is 15.
Thus, g = 13 - 15 = -2
h is the residual value where the given value is 20 and the predicted value is 19.
Thus, h = 20 - 19 = 1
Therefore, g = –2 and h = 1