Summary: In a forecasting problem, we have $\mathcal P$, the priors, e.g., price and demand is negatively correlated, $\mathcal D$, available dataset, $Y$, the observations, and $F$, the forecasts. Information Set $\mathcal A$ The priors $\mathcal D$ and the available data $\mathcal P$ can be summarized together as the information set $\mathcal A$. Under a probabilistic view, a forecaster will find out or approximate a CDF $\mathcal F$ such that1 $$ \mathcal F(Y\vert \mathcal D, \mathcal P) \to F. $$ Naively speaking, once the density $\rho(F, Y)$ is determined or estimated, a probabilistic forecaster can be formed. The joint probability of $(F, Y)$ is our prediction space.

Pages: 4

Time Series Forecasting

Knowledge snippets about forecasting

Introduction: My Knowledge Cards

⁵ Graph Structure Learning in GNN

⁴ Mix-hop Propagation in GNN

³ Time Convolution

¹ Prediction Space in Forecasting

Time Series Forecasting

Knowledge snippets about forecasting

Introduction: My Knowledge Cards

5 Graph Structure Learning in GNN

4 Mix-hop Propagation in GNN

3 Time Convolution

1 Prediction Space in Forecasting

⁵ Graph Structure Learning in GNN

⁴ Mix-hop Propagation in GNN

³ Time Convolution

¹ Prediction Space in Forecasting