Complete Revised SIR Model Forecasts (8 April): USA and Canada

in Data Science

on

Here is a full look at the outputs from our revised SIR model. I have included plots from hot spots in both the US and Canada as well as cities where I have friends, family, and loved ones. I can run these simulations for just about any city in the world, so if you have any cities you want to see, leave me a message in the comments or contact me directly.

Overview of SIR Model Output

Each city has four plots. The top row is the “working” model output, with the model curve best fit to the actual data. The bottom row is an experimental model output showing the effect of social distancing. In the “working” model runs on the top row, there are 5 lines on each plot. The middle line is the R Naught value that was reverse-engineered by fitting the model output to the actual data, and there are two lines on each side of the best-fit line showing different R Naught values in steps of 0.2.

Note: The y-axis on some of the experimental social distancing plots showing the total case count (bottom right plot for each city) is mislabeled. It should read “Total Cases”, not “Number of Infected”.

Finally, don’t forget that the plots below assume the R Naught values and the amount of social distancing remains constant throughout the entire time series. In reality, additional social distancing restrictions will dampen the curve and shift it to the right, while removing social distancing restrictions will cause the curve to accelerate and shift to the left.

Confidence in SIR Model Predictions

My confidence level in the “working”/top row model outputs is as follows:

  • Predicting the apex of the outbreak: medium-high to high. The curves should at least be “in the ballpark.”
  • Predicting the total number of cases: low to very low. With how fast things are changing right now and how fast new data is coming in, we just don’t know at this point. My gut feeling is that the case count projections in these model runs are likely high overall, but from a public health perspective, I would much rather have the model overestimate case counts than underestimate them.

Plots are in alphabetical order by city, with a table of additional cities at the bottom. Click on any plot to view it full size.

Boston, Massachusetts

  • Boston, MA SIR Model Number of Active Infections: Constant R Naught
  • Boston, MA SIR Model Total Cases: Constant R Naught
  • Boston, MA Number of Active Infections: Percent of People Social Distancing
  • Boston, MA Total Cases: Percent of People Social Distancing

Chicago, Illinois

  • Chicago, IL SIR Model Number of Active Infections: Constant R Naught
  • Chicago, IL SIR Model Total Cases: Constant R Naught
  • Chicago, IL Number of Active Infections: Percent of People Social Distancing
  • Chicago, IL Total Cases: Percent of People Social Distancing

Detroit, Michigan

  • Detroit, MI SIR Model Number of Active Infections: Constant R Naught
  • Detroit, MI SIR Model Total Cases: Constant R Naught
  • Detroit, MI Number of Active Infections: Percent of People Social Distancing
  • Detroit, MI Total Cases: Percent of People Social Distancing

Los Angeles, California

  • Los Angeles, CA SIR Model Number of Active Infections: Constant R Naught
  • Los Angeles, CA SIR Model Total Cases: Constant R Naught
  • Los Angeles, CA Number of Active Infections: Percent of People Social Distancing
  • Los Angeles, CA Total Cases: Percent of People Social Distancing

Montréal, Québec

  • Montreal, QC SIR Model Number of Active Infections: Constant R Naught
  • Montreal, QC SIR Model Total Cases: Constant R Naught
  • Montreal, QC Number of Active Infections: Percent of People Social Distancing
  • Montreal, QC Total Cases: Percent of People Social Distancing

New Orleans, Louisiana

  • New Orleans, LA SIR Model Number of Active Infections: Constant R Naught
  • New Orleans, LA SIR Model Total Cases: Constant R Naught
  • New Orleans, LA Number of Active Infections: Percent of People Social Distancing
  • New Orleans, LA Total Cases: Percent of People Social Distancing

New York, New York

  • New York, NY SIR Model Number of Active Infections: Constant R Naught
  • New York, NY SIR Model Total Cases: Constant R Naught
  • New York, NY Number of Active Infections: Percent of People Social Distancing
  • New York, NY Total Cases: Percent of People Social Distancing

Oklahoma City, Oklahoma

  • Oklahoma City, OK SIR Model Number of Active Infections: Constant R Naught
  • Oklahoma City, OK SIR Model Total Cases: Constant R Naught
  • Oklahoma City, OK Number of Active Infections: Percent of People Social Distancing
  • Oklahoma City, OK Total Cases: Percent of People Social Distancing

Ottawa, Ontario

  • Ottawa, ON SIR Model Number of Active Infections: Constant R Naught
  • Ottawa, ON SIR Model Total Cases: Constant R Naught
  • Ottawa, ON Number of Active Infections: Percent of People Social Distancing
  • Ottawa, ON Total Cases: Percent of People Social Distancing

Portland, Oregon

  • Portland, OR SIR Model Number of Active Infections: Constant R Naught
  • Portland, OR SIR Model Total Cases: Constant R Naught
  • Portland, OR Number of Active Infections: Percent of Population Social Distancing
  • Portland, OR Total Cases: Percent of Population Social Distancing

Phoenix, Arizona

  • Phoenix, AZ SIR Model Number of Active Infections: Constant R Naught
  • Phoenix, AZ SIR Model Total Cases: Constant R Naught
  • Phoenix, AZ Number of Active Infections: Percent of People Social Distancing
  • Phoenix, AZ Total Cases: Percent of People Social Distancing

San Francisco, California

  • San Francisco, CA SIR Model Number of Active Infections: Constant R Naught
  • San Francisco, CA SIR Model Total Cases: Constant R Naught
  • San Francisco, CA Number of Active Infections: Percent of People Social Distancing
  • San Francisco, CA Total Cases: Percent of People Social Distancing

Tampa, Florida

  • Tampa, FL SIR Model Number of Active Infections: Constant R Naught
  • Tampa, FL SIR Model Total Cases: Constant R Naught
  • Tampa, FL Number of Active Infections: Percent of People Social Distancing
  • Tampa, FL Total Cases: Percent of People Social Distancing

Toronto, Ontario

  • Toronto, ON SIR Model Number of Active Infections: Constant R Naught
  • Toronto, ON SIR Model Total Cases: Constant R Naught
  • Toronto, ON Number of Active Cases: Percent of People Social Distancing
  • Toronto, ON Total Cases: Percent of People Social Distancing

SIR Model Outputs for Additional Cities

Please note that this table contains outputs of just this single model run and does not necessarily reflect what my actual predictions are. I will be putting this table on my COVID-19 Pandemic Tracker later this week and regularly updating it there.

Data points I’m skeptical of in this output (with some comments):

  • Chicago, IL: Case count is likely overestimated. I’m not sure why, but the most likely reason is good social distancing.
  • Los Angeles, CA: Case count is likely overestimated due to California being better at social distancing than what was input into the model
  • Seattle, WA: Peak date is incorrect due to the State of Washington’s 100th case occurring before John’s Hopkins began breaking down data by state.
  • Washington, DC: Not enough data to accurately fit the curve
  • Winnipeg, MB: Not enough data to accurately fit the curve
CityState or ProvinceApex DateTotal Cases @ ApexInfected @ Apex
AtlantaGeorgiaLate April to Early May10,000 to 100,00010,000 to 100,000
BostonMassachusettsLate April to Early May50,000 to 200,00010,000 to 50,000
CalgaryAlbertaEarly June10,000 to 100,00010,000 to 50,000
ChicagoIllinoisMid-to-Late April100,000 to 500,000100,000 to 200,000
DallasTexasEarly May100,000 to 500,00050,000 to 100,000
DenverColoradoEarly-to-Mid May10,000 to 100,00010,000 to 50,000
DetroitMichiganMid-to-Late April50,000 to 100,00010,000 to 100,000
EdmontonAlbertaLate May to Early June10,000 to 100,00010,000 to 50,000
HoustonTexasEarly May100,000 to 500,00050,000 to 150,000
Los AngelesCaliforniaEarly May100,000 to 1,000,000100,000 to 500,000
MiamiFloridaLate April10,000 to 100,00010,000 to 50,000
MontréalQuébecLate April to Early May100,000 to 500,00010,000 to 100,000
New OrleansLouisianaMid-to-Late April10,000 to 100,00010,000 to 50,000
New YorkNew YorkMid-April100,000 to 1,000,000100,000 to 700,000
Oklahoma CityOklahomaEarly-to-Mid May10,000 to 100,00010,000 to 50,000
OttawaOntarioMid May50,000 to 200,00010,000 to 50,000
PhiladelphiaPennsylvaniaLate April to Early May50,000 to 500,00050,000 to 100,000
PhoenixArizonaMid May10,000 to 200,00010,000 to 100,000
PortlandOregonLate May to Early June10,000 to 100,0005,000 to 50,000
SeattleWashingtonLate April to Early May10,000 to 100,00010,000 to 50,000
San FranciscoCaliforniaLate April to Early May50,000 to 200,00010,000 to 50,000
TampaFloridaMid-to-Late April10,000 to 100,00010,000 to 50,000
TorontoOntarioMid-to-Late May100,000 to 500,00050,000 to 200,000
VancouverBritish ColumbiaEarly to Mid June10,000 to 100,0005,000 to 50,000
WashingtonDistrict of ColumbiaLate May to Early June10,000 to 100,00010,000 to 50,000
WinnipegManitobaLate June to Early July10,000 to 100,0001,000 to 20,000

Comments are closed.