Economic Nowcasting with Long Short-term Memory Artificial Neural Networks (LSTM)

Hopp, Daniel

doi:10.2139/ssrn.3855402

Cited by 11 publications

(15 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some of the most common include the dynamic factor model (DFM) [76,77], mixed data sampling (MIDAS) [45,78], mixed data sampling vector autoregression (VAR) [78], and Bayesian vector autoregression [79]. Hopp [80] examined an approach using long short-term memory artificial neural networks (LSTM).…”

Section: Methodsmentioning

confidence: 99%

Feasibility of nowcasting SDG indicators: A comprehensive survey

Hopp¹,

Fu²,

Peltola³

2022

SJI

Self Cite

View full text Add to dashboard Cite

The 2030 Agenda and accompanying Sustainable Development Goals (SDGs) are vital in guiding national and global policy. However, many of the SDG indicators used to measure progress toward those goals suffer from long publication lags. Nowcasting has the potential to address this problem and generate more timely estimates of those indicators. This paper provides resources for achieving that potential by 1) carrying out a comprehensive nowcasting feasibility survey of all SDG indicators to assess their potential to be nowcast, and 2) performing a case study of indicator 9.4.1 to illustrate and shed light on the process of performing a nowcasting exercise. There exist 231 SDG indicators, but due to only examining Tier 1 indicators and the fact that many indicators have multiple sub-indicators, 362 indicators and sub-indicators were eventually surveyed. Of those 362, 150 were found highly likely to be suitable candidates for nowcasting, 87 were found to be likely, and 125 were found to be unsuitable.

show abstract

Section: Methodsmentioning

confidence: 99%

Feasibility of nowcasting SDG indicators: A comprehensive survey

Hopp¹,

Fu²,

Peltola³

2022

SJI

Self Cite

View full text Add to dashboard Cite

show abstract

“…Some common methodologies to perform economic nowcasting include mixed data sampling (MIDAS) [14,15], dynamic factor models (DFM) [13,16], mixed-frequency vector autoregression (VAR) [14,17], and Bayesian vector autoregressions [18]. Hopp [3] and Loermann and Maas [19] examined neural networks' suitability to the application, more specifically long short-term memory (LSTM) networks in the case of the former. The LSTM methodology is explained further in Section 2.2.…”

Section: Nowcasting In the Context Of The Covid-19 Pandemicmentioning

confidence: 99%

“…The LSTM methodology is explained further in Section 2.2. For more information on nowcasting, including commentary on common data issues encountered in the field, see Hopp [3] or Cimadomo et al [18].…”

Section: Nowcasting In the Context Of The Covid-19 Pandemicmentioning

confidence: 99%

“…This allows them to better handle time series and renders them more suitable for application in the nowcasting context [28]. For more information on how ANNs and LSTMs work, see: Hopp [3], Singh and Prajneshu [29], Sazli [30], or Loermann and Maas [19]. For more detailed information on LSTMs' properties which make it suitable for nowcasting, see Hopp [3], Section 3.2.…”

Section: Long Short-term Memory Neural Network and Dynamic Factor Modelsmentioning

confidence: 99%

“…These circumstances make 2020 a particularly interesting case in which to examine the performance of different nowcasting methodologies. This paper seeks to do just that, assessing two methodologies, the dynamic factor model (DFM), currently a popular choice in economic nowcasting, and the long short-term memory neural network (LSTM), explored in-depth in a previous United Nations Conference on Trade and Development (UNC-TAD) research paper [3]. Both the DFM and LSTM methodologies employed for the analysis are available as open-source libraries; in R in the case of the DFM [4] and in Python, R, MATLAB, and Julia in the case of the LSTM [5][6][7][8].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Performance of LSTM neural networks in nowcasting global trade during the COVID-19 crisis

Hopp¹

2022

SJI

Self Cite

View full text Add to dashboard Cite

The COVID-19 pandemic has demonstrated the increasing need of policymakers for timely estimates of macroeconomic variables. A prior United Nations Conference on Trade and Development (UNCTAD) research paper examined the suitability of long short-term memory artificial neural networks (LSTM) for performing economic nowcasting of this nature. Building off those findings, in this paper, the LSTM’s performance during the COVID-19 pandemic is compared and contrasted with that of the dynamic factor model (DFM), a commonly used methodology in the field. Three separate variables, global merchandise export values and volumes and global services exports, were nowcast with actual data vintages and performance evaluated for the second, third, and fourth quarters of 2020 and the first and second quarters of 2021. In terms of both mean absolute error and root mean square error, the LSTM obtained better performance in two-thirds of variable/quarter combinations, as well as displayed more gradual forecast evolutions with more consistent narratives and smaller revisions.

show abstract

Making the hospital smart: using a deep long short-term memory model to predict hospital performance metrics

Jia

Zhu

et al. 2022

IMDS

View full text Add to dashboard Cite

PurposeAbundant studies of outpatient visits apply traditional recurrent neural network (RNN) approaches; more recent methods, such as the deep long short-term memory (DLSTM) model, have yet to be implemented in efforts to forecast key hospital data. Therefore, the current study aims to reports on an application of the DLSTM model to forecast multiple streams of healthcare data.Design/methodology/approachAs the most advanced machine learning (ML) method, static and dynamic DLSTM models aim to forecast time-series data, such as daily patient visits. With a comparative analysis conducted in a high-level, urban Chinese hospital, this study tests the proposed DLSTM model against several widely used time-series analyses as reference models.FindingsThe empirical results show that the static DLSTM approach outperforms seasonal autoregressive integrated moving averages (SARIMA), single and multiple RNN, deep gated recurrent units (DGRU), traditional long short-term memory (LSTM) and dynamic DLSTM, with smaller mean absolute, root mean square, mean absolute percentage and root mean square percentage errors (RMSPE). In particular, static DLSTM outperforms all other models for predicting daily patient visits, the number of daily medical examinations and prescriptions.Practical implicationsWith these results, hospitals can achieve more precise predictions of outpatient visits, medical examinations and prescriptions, which can inform hospitals' construction plans and increase the efficiency with which the hospitals manage relevant information.Originality/valueTo address a persistent gap in smart hospital and ML literature, this study offers evidence of the best forecasting models with a comparative analysis. The study extends predictive methods for forecasting patient visits, medical examinations and prescriptions and advances insights into smart hospitals by testing a state-of-the-art, deep learning neural network method.

show abstract

Economic Nowcasting with Long Short-term Memory Artificial Neural Networks (LSTM)

Cited by 11 publications

References 27 publications

Feasibility of nowcasting SDG indicators: A comprehensive survey

Feasibility of nowcasting SDG indicators: A comprehensive survey

Performance of LSTM neural networks in nowcasting global trade during the COVID-19 crisis

Making the hospital smart: using a deep long short-term memory model to predict hospital performance metrics

Contact Info

Product

Resources

About