The state of the art in probabilistic demand forecasting [40] minimizes Quantile Loss to predict the future demand quantiles for different horizons. However, since quantiles aren’t additive, in order to predict the total demand for any wider future interval all required intervals are usually appended to the target vector during model training. The separate optimization of these overlapping intervals can lead to inconsistent forecasts, i.e. forecasts which imply an invalid joint distribution between different horizons. As a result, inter-temporal decision making algorithms that depend on the joint or step-wise conditional distribution of future demand cannot utilize these forecasts. In this work, we address the problem by using sample paths to predict future demand quantiles in a consistent manner and propose several novel methodologies to solve this problem. Our work covers the use of covariance shrinkage methods, autoregressive models, generative adversarial networks and also touches on the use of variational autoencoders and Bayesian Dropout.