Hello Everyone,
My research team is trying to develop an COVID-19 prediction model using time series data. However, we are unable to get the dataset. It is kindly requested to share or provide a link for an available public dataset.
Thanks in Advance.
the best websites are these two:
our world in data:
https://ourworldindata.org/coronavirus#licence
(data in an excel sheet)
worldometer:
https://www.worldometers.info/coronavirus/?fbclid=IwAR0zAnmqgism07DU3q2p2BTu6AzRjnB8CoXVnY0Un6SfuLUrl6kZP3VM8Sw#countries
(gathering data through plots)
I would suggest contacting the Statisics Group at St. Jude's Children's Research Hospital ln Memphis, TN. They had a great webinar conference on this topic and I believe the talk videos and notes are still posted on the internet. I attended the conference which was excellent. I suggest however that there are much better ways to model such data than pure time series models which go back to about 1900.in epidemiology. The webinar discussed this approach in some detail. Really good conference.the group email should be easy to find on the.internet
Best wishes, David Booth
I believe that your research team is going to fail in trying to develop a practically useful COVID-19 prediction model using only time series data cases.
Indeed, Covid19 cases prediction models are usually based on (i) SEIR/SIR (susceptible- infected-recovered) set of differential equations, or (ii) agent-based simulations, or (iii) extrapolation based on time series, (iv) and/or curve-fitting approaches (logistic type models, ARIMA, etc).
However, all models include a number of adjustable parameters and variables. Every variable is dependent on a number of assumptions and the other variables. Despite a lot of efforts worldwide, there are significant knowledge gaps. Different countries and regions collect data in different ways. There’s no single standard form to fill in that can easily allow comparing cases around the world.
The same inconsistencies apply to who gets tested. Some countries are giving tests to anyone who wants one. Others are not. That affects how much is known about the number of people who have actually contracted COVID-19, versus how many people have tested positive depending on the test nature.
And the virus itself is an unpredictable contagion, hurting some groups more than others — meaning that local demographics and health care access are going to be big determinants when it comes to the virus’s impact on communities.
Major uncertainties come from multiple different sources, such as demographics, comorbidity that can be lung-related or heart-related diseases, diabetes prevalence, hospital settings/capacity, test rate, social distancing/habits, and income levels. No existing forecasting models of various types barely include the aforementioned factors. And the models' outcome is very sensitive to the numerical values of the included parameters, and other structural assumptions.
On top of that, forecasting the number of cases itself is of little practical value. What is needed is information on the factors that can be leveraged to reduce the number of cases. These factors can include both individual susceptibility/immunity to the infection and the virus genetic mutational behaviors. These factors are almost impossible to formalize to include in the accurate forecasting models, they are mostly out of control of the policy decision-makers.
The bottom line:
(i) in epidemiology, the general types and structures of theoretical mathematical forecasting models of an epidemic spread are well established.
(ii) however, filling the models with accurate dynamic information is still a huge challenge.
(iii) no theoretical model of a pandemic spread and its rate is reliable and accurate enough in the long run yet.
(iv) the empirical information of the number of cases and the ways of its collection in different areas are inconsistent and difficult to compare to each other in order to become a reliable help for theoretical models.
I agree with Prof. Alexander Kolker's submissions. The most difficult part of this is that the virus is unpredictable
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication