The links to the data do not explicitely tell at which frequencies they are updated. This particurlarly refers to the datasets aggregated to monthly (or in some cases even to 3-month) means. Would you indicate on the same page the latency for each suggested dataset?
In general, you should use the linked data sources to determine expected publishing latency. Furthermore, the data sources in general are not under the control of the challenge organizers, so there is no guarantee that the latency will be consistent. Additionally, some of the data sources may publish provisional data that later gets updated. This is a limitation of using real-world operational data to build models for operational forecasts. You may want to take into consideration when writing the code for your model (e.g., having it be able to work with whatever data is the latest available).
Let us know if you have any further questions.
Hi @jayqi ,
Thank you for the clarifications. I now understand the circumstances and appreciate your suggestion on handling this uncertainty in the code.