Back

Forecast Intervals for Infectious Disease Models

Picard, R.; Osthus, D.

2022-05-01 epidemiology
10.1101/2022.04.29.22274494 medRxiv
Show abstract

Forecast intervals for infectious disease transmission and mortality have long been overconfident -- i.e., the advertised coverage probabilities of those intervals fell short of their subsequent performances. Further, there was no apparent relation between how good models claimed to be (as measured by their purported forecast uncertainties) and how good the models really were (as measured by their actual forecast errors). The main cause of this problem lies in the misapplication of textbook methods for uncertainty quantification. A solution lies in the creative use of predictive tail probabilities to obtain valid interval coverages. This approach is compatible with all probabilistic predictive models whose forecast error behavior does not change "too quickly" over time.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

1
Epidemics
○ 116 papers in training set
Top 0.1%
15.2%
Similar papers in this journal
2
PLOS Computational Biology
● 1863 papers in training set
Top 3%
9.9%
Similar papers in this journal
3
PLOS ONE
● 5266 papers in training set
Top 23%
6.8%
Similar papers in this journal
4
Mathematical Biosciences
○ 49 papers in training set
Top 0.1%
5.5%
Similar papers in this journal
5
Infectious Disease Modelling
○ 54 papers in training set
Top 0.3%
4.9%
Similar papers in this journal
6
Journal of The Royal Society Interface
● 235 papers in training set
Top 0.8%
4.9%
7
Epidemiology
○ 32 papers in training set
Top 0.1%
3.3%
50% of probability mass above
8
Journal of Theoretical Biology
○ 162 papers in training set
Top 0.9%
3.3%
9
Statistics in Medicine
○ 40 papers in training set
Top 0.2%
3.2%
10
American Journal of Epidemiology
◐ 67 papers in training set
Top 0.4%
2.7%
11
eLife
● 5828 papers in training set
Top 41%
2.5%
12
Medical Decision Making
○ 12 papers in training set
Top 0.1%
2.5%
13
Frontiers in Public Health
○ 148 papers in training set
Top 2%
2.5%
14
PeerJ
◐ 308 papers in training set
Top 4%
1.9%
15
Scientific Reports
○ 3612 papers in training set
Top 51%
1.7%
16
Frontiers in Artificial Intelligence
○ 20 papers in training set
Top 0.3%
1.7%
17
Statistical Methods in Medical Research
○ 11 papers in training set
Top 0.1%
1.7%
18
Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
● 12 papers in training set
Top 0.1%
1.3%
19
BMC Medical Research Methodology
○ 47 papers in training set
Top 0.9%
1.1%
20
Bulletin of Mathematical Biology
○ 92 papers in training set
Top 1%
1.1%
21
Mathematical Medicine and Biology: A Journal of the IMA
◐ 10 papers in training set
Top 0.1%
1.1%
22
Royal Society Open Science
● 214 papers in training set
Top 4%
1.1%
23
Mathematical Biosciences and Engineering
● 23 papers in training set
Top 0.7%
1.1%
24
Physical Biology
○ 46 papers in training set
Top 0.7%
1.0%
25
Epidemiology and Infection
◐ 89 papers in training set
Top 3%
0.9%
26
Biometrics
◐ 23 papers in training set
Top 0.3%
0.9%
27
Emerging Infectious Diseases
● 105 papers in training set
Top 2%
0.6%
28
Computer Methods in Biomechanics and Biomedical Engineering
○ 10 papers in training set
Top 0.3%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.