ATM cash demand forecasting in an Indian bank with chaos and hybrid deep learning networks

作者：

Highlights：

• Deep learning models, LSTM, 1D-CNN outperformed other models w.r.t SMAPE.

• RF, LSTM & 1D-CNN + LSTM are statistically similar in non-chaos-observed-ATMs.

• In chaos-observed-ATMs, RF and LSTM was statistically significant.

• GRU displayed less variability in predictions than other techniques w.r.t. SMAPE.

• Including exogenous dummy variable significantly improved the predictions.

摘要

•Deep learning models, LSTM, 1D-CNN outperformed other models w.r.t SMAPE.•RF, LSTM & 1D-CNN + LSTM are statistically similar in non-chaos-observed-ATMs.•In chaos-observed-ATMs, RF and LSTM was statistically significant.•GRU displayed less variability in predictions than other techniques w.r.t. SMAPE.•Including exogenous dummy variable significantly improved the predictions.

论文关键词：ATM Cash Demand Forecasting,Deep Learning,Chaos,1D-CNN,LSTM,GRNN,GMDH

论文评审过程：Received 31 December 2021, Revised 17 August 2022, Accepted 19 August 2022, Available online 23 August 2022, Version of Record 29 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118645