Gaussian process nowcasting: application to COVID-19 mortality reporting

Iwona Hawryluk, Henrique Hoeltgebaum, Swapnil Mishra, Xenia Miscouridou, Ricardo P Schnekenberg, Charles Whittaker, Michaela Vollmer, Seth Flaxman, Samir Bhatt, Thomas A. Mellan
Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence, PMLR 161:1258-1268, 2021.

Abstract

Updating observations of a signal due to the delays in the measurement process is a common problem in signal processing, with prominent examples in a wide range of fields. An important example of this problem is the nowcasting of COVID-19 mortality: given a stream of reported counts of daily deaths, can we correct for the delays in reporting to paint an accurate picture of the present, with uncertainty? Without this correction, raw data will often mislead by suggesting an improving situation. We present a flexible approach using a latent Gaussian process that is capable of describing the changing auto-correlation structure present in the reporting time-delay surface. This approach also yields robust estimates of uncertainty for the estimated nowcasted numbers of deaths. We test assumptions in model specification such as the choice of kernel or hyper priors, and evaluate model performance on a challenging real dataset from Brazil. Our experiments show that Gaussian process nowcasting performs favourably against both comparable methods, and against a small sample of expert human predictions. Our approach has substantial practical utility in disease modelling — by applying our approach to COVID-19 mortality data from Brazil, where reporting delays are large, we can make informative predictions on important epidemiological quantities such as the current effective reproduction number.

Cite this Paper

BibTeX
@InProceedings{pmlr-v161-hawryluk21a, title = {Gaussian process nowcasting: application to COVID-19 mortality reporting}, author = {Hawryluk, Iwona and Hoeltgebaum, Henrique and Mishra, Swapnil and Miscouridou, Xenia and Schnekenberg, Ricardo P and Whittaker, Charles and Vollmer, Michaela and Flaxman, Seth and Bhatt, Samir and Mellan, Thomas A.}, booktitle = {Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence}, pages = {1258--1268}, year = {2021}, editor = {de Campos, Cassio and Maathuis, Marloes H.}, volume = {161}, series = {Proceedings of Machine Learning Research}, month = {27--30 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v161/hawryluk21a/hawryluk21a.pdf}, url = {https://proceedings.mlr.press/v161/hawryluk21a.html}, abstract = {Updating observations of a signal due to the delays in the measurement process is a common problem in signal processing, with prominent examples in a wide range of fields. An important example of this problem is the nowcasting of COVID-19 mortality: given a stream of reported counts of daily deaths, can we correct for the delays in reporting to paint an accurate picture of the present, with uncertainty? Without this correction, raw data will often mislead by suggesting an improving situation. We present a flexible approach using a latent Gaussian process that is capable of describing the changing auto-correlation structure present in the reporting time-delay surface. This approach also yields robust estimates of uncertainty for the estimated nowcasted numbers of deaths. We test assumptions in model specification such as the choice of kernel or hyper priors, and evaluate model performance on a challenging real dataset from Brazil. Our experiments show that Gaussian process nowcasting performs favourably against both comparable methods, and against a small sample of expert human predictions. Our approach has substantial practical utility in disease modelling — by applying our approach to COVID-19 mortality data from Brazil, where reporting delays are large, we can make informative predictions on important epidemiological quantities such as the current effective reproduction number.} }
Endnote
%0 Conference Paper %T Gaussian process nowcasting: application to COVID-19 mortality reporting %A Iwona Hawryluk %A Henrique Hoeltgebaum %A Swapnil Mishra %A Xenia Miscouridou %A Ricardo P Schnekenberg %A Charles Whittaker %A Michaela Vollmer %A Seth Flaxman %A Samir Bhatt %A Thomas A. Mellan %B Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2021 %E Cassio de Campos %E Marloes H. Maathuis %F pmlr-v161-hawryluk21a %I PMLR %P 1258--1268 %U https://proceedings.mlr.press/v161/hawryluk21a.html %V 161 %X Updating observations of a signal due to the delays in the measurement process is a common problem in signal processing, with prominent examples in a wide range of fields. An important example of this problem is the nowcasting of COVID-19 mortality: given a stream of reported counts of daily deaths, can we correct for the delays in reporting to paint an accurate picture of the present, with uncertainty? Without this correction, raw data will often mislead by suggesting an improving situation. We present a flexible approach using a latent Gaussian process that is capable of describing the changing auto-correlation structure present in the reporting time-delay surface. This approach also yields robust estimates of uncertainty for the estimated nowcasted numbers of deaths. We test assumptions in model specification such as the choice of kernel or hyper priors, and evaluate model performance on a challenging real dataset from Brazil. Our experiments show that Gaussian process nowcasting performs favourably against both comparable methods, and against a small sample of expert human predictions. Our approach has substantial practical utility in disease modelling — by applying our approach to COVID-19 mortality data from Brazil, where reporting delays are large, we can make informative predictions on important epidemiological quantities such as the current effective reproduction number.
APA
Hawryluk, I., Hoeltgebaum, H., Mishra, S., Miscouridou, X., Schnekenberg, R.P., Whittaker, C., Vollmer, M., Flaxman, S., Bhatt, S. & Mellan, T.A.. (2021). Gaussian process nowcasting: application to COVID-19 mortality reporting. Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 161:1258-1268 Available from https://proceedings.mlr.press/v161/hawryluk21a.html.

Related Material