An Auto Regressive Deep Learning Model for Sales Tax Forecasting from Multiple Short Time Series

Boosting tax revenues with mixed-frequency data in the aftermath of COVID-19: The case of New York

“…But these powerful methods have not been widely used in governmental budget forecasting. An exception is Buxton et al (2019), who used deep learning to forecast the sales tax revenues in Illinois. One of the weaknesses of existing literature is that the importance of business cycles is not fully integrated in budget forecasting.…”