Abstract:
|
For any financial organization, computing accurate quarterly forecasts is a critical operation. As the granularity at which forecasts are needed increases, traditional statistical time series models may not scale well. We apply deep neural networks in the forecasting domain by experimenting with techniques from Natural Language Processing (Encoder-Decoder LSTMs) and Computer Vision (Dilated CNNs), as well as incorporating transfer learning. We illustrate model performance using Microsoft’s revenue data corresponding to Enterprise, and Small, Medium & Corporate products, spanning approximately 60 regions worldwide for 8 different business segments, and totaling tens of billions of USD. We compare our models’ performance to the ensemble model (of traditional statistics and machine learning) currently being used by Microsoft Finance. Using this in-production model as a baseline, our experiments yield an approximately 30% improvement overall in accuracy on test data, with the most substantial gains arising from our novel application of curriculum learning to time series forecasting.
|