Mark the model for Sweden as invalid or similar

[DrChr]

Due to the model for Sweden using incorrect data, I think you should somehow mark the corresponding web page [1] as invalid or something. Perhaps you could add a banner or note at the top of the web page. Or maybe just remove it.

As for why the data I say the data it's using is invalid, have a look at the the screenshots below for comparison. See #58 for additional details.

I also wonder if you have checked the data for other countries?

N:o deceased for different days as taken from Sweden's "Folkhälsomyndighet" on the 16th of April:

Compare this with what's currently (on the 16th) being shown on the model for Sweden:

[1] https://mrc-ide.github.io/covid19estimates/#/details/Sweden

[DrChr]

Below is a comparison of data from Folkhälsomyndigheten (FOHM) and data I just downloaded from ECDC's web site:

Date	FOHM	ECDC	Error
3/11/2020	1	0	1
3/12/2020	0	1	-1
3/13/2020	1	0	1
3/14/2020	1	0	1
3/15/2020	2	0	2
3/16/2020	2	2	0
3/17/2020	1	4	-3
3/18/2020	6	1	5
3/19/2020	7	2	5
3/20/2020	10	0	10
3/21/2020	7	6	1
3/22/2020	12	4	8
3/23/2020	11	1	10
3/24/2020	20	4	16
3/25/2020	25	11	14
3/26/2020	30	6	24
3/27/2020	32	24	8
3/28/2020	35	26	9
3/29/2020	38	10	28
3/30/2020	42	8	34
3/31/2020	43	36	7
4/1/2020	50	34	16
4/2/2020	68	59	9
4/3/2020	71	43	28
4/4/2020	61	51	10
4/5/2020	79	40	39
4/6/2020	85	28	57
4/7/2020	75	76	-1
4/8/2020	97	114	-17
4/9/2020	63	96	-33
4/10/2020	62	106	-44
4/11/2020	61	77	-16
4/12/2020	62	17	45
4/13/2020	55	12	43
4/14/2020	49	20	29
4/15/2020	41	114	-73
4/16/2020	10	170	-152
Date missing:	18

[offroff]

The data from the swedish authorities are quite problematic, I wouldn't use it. Also, the swedish data has fooled the authority itself and the swedish people. The ECDC data is probably a lot more useful.