The impact of forecasting error in wind power on unscheduled flows (USFs) is investigated here. Normal distribution is used to model the forecasting error distribution. Upper and lower bounds on wind farm output with a positive correlation of errors are obtained. Monte Carlo simulations using the interval forecasts of wind farm outputs are run to obtain interval branch flows. Ordinary least squares and ridge regression are used for the estimation of a mathematical artifact - minor loop flows - for accommodating USFs. Model adequacy and statistical inferences of the loop flow estimates is discussed. Impact of forecasting error on distributions of estimated loop flow is explored on the basis of Kolmogorov-Smirnov (KS) and chi-square goodness-of-fit tests.