Project: CDC Real-time Forecasts API Config

Models (39)

• Augmented Mechanistic GLEAM Ensemble
• Auto Regressive model with Likelihood Ratio based Model Selection
• Basis Regression (epiforecast package defaults)
• Bayesian Model Averaging
• Delta Density (epiforecast package defaults)
• Dynamic Bayesian Model plus
• Dynamic Bayesian Model with a hierarchical discrepancy, 2017/2018 version
• Dynamic Harmonic Model with ARIMA errors
• Empirical Bayes (conditioning on past four weeks only, 2017/2018 version)
• Empirical Bayes (epiforecast package defaults, 2017/2018 version)
• Empirical Futures (epiforecast package defaults)
• Empirical Trajectories (epiforecast package defaults)
• Ensemble Adjustment Kalman Filter SEIRS
• Ensemble Adjustment Kalman Filter SIRS
• Ensemble Kalman Filter SEIRS
• Ensemble Kalman Filter SIRS
• Ensemble of dynamic harmonic model and historical averages
• Epidemic Cosine
• Kernel Conditional Density Estimation
• Kernel Conditional Density Estimation, post-hoc backfill adjustment
• Kernel Density Estimation
• KoT-adaptive
• Markovian Delta Density (epiforecast package defaults)
• Mechanistic GLEAM Ensemble
• Rank Histogram Filter SEIRS
• Rank Histogram Filter SIRS
• Recurrent Neural Network (Long Short-Term Memory)
• SARIMA model with seasonal differencing
• SARIMA model without seasonal differencing
• Statistical Ensemble (using the eight submitted components, with no backcasting or nowcasting, 2017/2018 version)
• Subtype weighted historical average model
• UTA - Empirical dynamic model, 2017/2018 version
• Uniform Distribution
• constant-weights
• equal-weights
• static-regularized-target-region-weights
• target-region-weights
• target-type-weights
• target-weights

Locations (11)

HHS Region 1, HHS Region 10, HHS Region 2, HHS Region 3, HHS Region 4, HHS Region 5, HHS Region 6, HHS Region 7, HHS Region 8, HHS Region 9, US National

Targets (7)

Name	Description	Unit	Point Value Type	Date?	Step Ahead?	Predictions	BinLwrs
1 wk ahead	One- to four-week ahead forecasts will be defined as the weighted ILINet percentage for the target week.	percent	Float		1 step	BL|NM|PT|SA	131
2 wk ahead	One- to four-week ahead forecasts will be defined as the weighted ILINet percentage for the target week.	percent	Float		2 steps	BL|NM|PT|SA	131
3 wk ahead	One- to four-week ahead forecasts will be defined as the weighted ILINet percentage for the target week.	percent	Float		3 steps	BL|NM|PT|SA	131
4 wk ahead	One- to four-week ahead forecasts will be defined as the weighted ILINet percentage for the target week.	percent	Float		4 steps	BL|NM|PT|SA	131
Season onset	The onset of the season is defined as the MMWR surveillance week (http://wwwn.cdc.gov/nndss/script/downloads.aspx) when the percentage of visits for influenza-like illness (ILI) reported through ILINet reaches or exceeds the baseline value for three consecutive weeks (updated 2016-2017 ILINet baseline values for the US and each HHS region will be available at http://www.cdc.gov/flu/weekly/overview.htm the week of October 10, 2016). Forecasted 'onset' week values should be for the first week of that three week period.	week	Text			BC|BI|SC	0
Season peak percentage	The intensity will be defined as the highest numeric value that the weighted ILINet percentage reaches during the 2016-2017 influenza season.	percent	Float			BL|NM|PT|SA	131
Season peak week	The peak week will be defined as the MMWR surveillance week that the weighted ILINet percentage is the highest for the 2016-2017 influenza season.	week	Text			BC|BI|SC	0

Time zeros (92)