Is there anyone facing any problem with the John Hopkins coronavirus dataset?

Hi, is there anyone facing any problem with the John Hopkins coronavirus dataset? It seems likes, in some counties, the number of confirmed cases goes down in subsequent dates which should not happen. Is there any way to fix that? Is there any other dataset which will be reliable to use?

@Tanmoy_Bhowmik_University_of_Central_Florida I’d try these three sources: SafeGraph Community Data Products - Google Sheets, https://covidtracking.com/data, and
https://coronavirus.1point3acres.com/en

There are “declining” cumulative cases and deaths in most if not all available datasets. Forcing the data to be monotonically increasing would require making stronger assumptions than the data maintainers are not willing to make

Thank you for your comments. But in that case, if I want to know the number of new cases each day, what should I do regarding the negative cases? Do you have any suggestion?

I think different people have different solutions. The one I’ve used is to back-apply decreased values over earlier, higher values. That assumes that later reported values are more accurate than earlier reported values