A0327
Title: Forecast reconciliation using linear models: Study on time series with network structure
Authors: Mahsa Ashouri - Academia Sinica (Taiwan) [presenting]
Sadid Sahami - National Tsing Hua University (Taiwan)
Frederick Kin Hing Phoa - Academia Sinica (Taiwan)
Abstract: Forecasting hierarchical or grouped time series using a reconciliation approach involves two steps: computing base forecasts and reconciling the forecasts. Base forecasts can be computed by popular time series forecasting methods such as Exponential Smoothing (ETS) and Autoregressive Integrated Moving Average (ARIMA) models. The reconciliation step is a linear process that adjusts the base forecasts to ensure they are coherent. However, using ETS or ARIMA for base forecasts can be computationally challenging when there are a large number of series to forecast, as each model must be numerically optimized for each series. We propose a linear model that avoids this computational problem and handles the forecasting and reconciliation in a single step. This approach can also be extended to the network time series structure. This extended framework is used for forecasting if we have the network structure at each hierarchy level which applies the Least Absolute Shrinkage and Selection Operator (LASSO) to justify network connections. We illustrate our method using the export Free-on-Board (FOB) dataset.
