Trends
Reflecting this episode Nadieh Bremer and Zan Armstrong recommended avoiding aggregating time series data in different units than days, unless required. I thought about my last trend visualizations, where I considered it better to represent the sum of each month and display single values within each month per country. It might be that some interesting information could be missing.
So, I plotted again in a daily-basis.
I noticed that Sweden had extreme ups and downs so. With the help of this interactive plot, positioning the pointer over those interesting changes, I could see that there are days where Sweden reported zero cases during Saturday, Sunday, and Monday. That is, they only reported cases from Tuesday through Friday. This insight information was not captured by showing a monthly average.
From the information below, a similar situation, with a less clear pattern of the days, is present in the rest of the Nordic countries. Still, a monthly trend hid that information and biased summary statistics, in case some cases were attended but not adequately recorded.
zero_cases= nordic[ which(nordic$location=='Sweden'
& nordic$new_cases ==0), ]
#head(zero_cases)
table(zero_cases$date)##
## 2020-02-02 2020-02-03 2020-02-04 2020-02-05 2020-02-06 2020-02-07 2020-02-08
## 1 1 1 1 1 1 1
## 2020-02-09 2020-02-10 2020-02-11 2020-02-12 2020-02-13 2020-02-14 2020-02-15
## 1 1 1 1 1 1 1
## 2020-02-16 2020-02-17 2020-02-18 2020-02-19 2020-02-20 2020-02-21 2020-02-22
## 1 1 1 1 1 1 1
## 2020-02-23 2020-02-24 2020-02-25 2020-03-01 2020-08-28 2020-08-29 2020-08-30
## 1 1 1 1 1 1 1
## 2020-09-05 2020-09-06 2020-09-12 2020-09-13 2020-09-14 2020-09-19 2020-09-20
## 1 1 1 1 1 1 1
## 2020-09-21 2020-09-26 2020-09-27 2020-09-28 2020-10-03 2020-10-04 2020-10-05
## 1 1 1 1 1 1 1
## 2020-10-10 2020-10-11 2020-10-12 2020-10-17 2020-10-18 2020-10-19 2020-10-24
## 1 1 1 1 1 1 1
## 2020-10-25 2020-10-26 2020-10-31 2020-11-01 2020-11-02 2020-11-07 2020-11-08
## 1 1 1 1 1 1 1
## 2020-11-09 2020-11-14 2020-11-15 2020-11-16 2020-11-21 2020-11-22 2020-11-23
## 1 1 1 1 1 1 1
## 2020-11-28 2020-11-29 2020-11-30 2020-12-05 2020-12-06 2020-12-07 2020-12-12
## 1 1 1 1 1 1 1
## 2020-12-13 2020-12-14 2020-12-19 2020-12-20 2020-12-21 2020-12-24 2020-12-25
## 1 1 1 1 1 1 1
## 2020-12-26 2020-12-27 2020-12-28 2020-12-31 2021-01-01 2021-01-02 2021-01-03
## 1 1 1 1 1 1 1
## 2021-01-04 2021-01-06 2021-01-09 2021-01-10 2021-01-11 2021-01-16 2021-01-17
## 1 1 1 1 1 1 1
## 2021-01-18 2021-01-23 2021-01-24 2021-01-25 2021-01-30 2021-01-31 2021-02-01
## 1 1 1 1 1 1 1
## 2021-02-06 2021-02-07 2021-02-08 2021-02-13 2021-02-14 2021-02-15 2021-02-20
## 1 1 1 1 1 1 1
## 2021-02-21 2021-02-22 2021-02-27 2021-02-28 2021-03-01 2021-03-06 2021-03-07
## 1 1 1 1 1 1 1
## 2021-03-08 2021-03-13 2021-03-14 2021-03-15 2021-03-20 2021-03-21 2021-03-22
## 1 1 1 1 1 1 1zero_cases_ns= nordic[ which(nordic$location!='Sweden'
& nordic$new_cases ==0), ]
table(zero_cases_ns$date)##
## 2020-02-01 2020-02-02 2020-02-03 2020-02-04 2020-02-05 2020-02-06 2020-02-07
## 1 1 1 1 1 1 1
## 2020-02-08 2020-02-09 2020-02-10 2020-02-11 2020-02-12 2020-02-13 2020-02-14
## 1 1 1 1 1 1 1
## 2020-02-15 2020-02-16 2020-02-17 2020-02-18 2020-02-19 2020-02-20 2020-02-21
## 1 2 1 1 2 1 1
## 2020-02-22 2020-02-23 2020-02-24 2020-02-25 2020-02-27 2020-02-28 2020-02-29
## 1 1 1 1 2 2 1
## 2020-03-02 2020-03-03 2020-03-04 2020-03-05 2020-03-07 2020-03-08 2020-03-12
## 2 1 1 1 2 1 1
## 2020-04-08 2020-04-24 2020-04-27 2020-04-30 2020-05-02 2020-05-04 2020-05-05
## 1 1 1 1 1 1 1
## 2020-05-06 2020-05-08 2020-05-09 2020-05-10 2020-05-11 2020-05-12 2020-05-14
## 1 1 1 1 1 1 1
## 2020-05-15 2020-05-16 2020-05-17 2020-05-18 2020-05-19 2020-05-21 2020-05-22
## 1 1 1 1 1 1 1
## 2020-05-24 2020-05-25 2020-05-26 2020-05-28 2020-05-29 2020-05-31 2020-06-01
## 1 1 1 1 1 1 1
## 2020-06-02 2020-06-03 2020-06-04 2020-06-05 2020-06-06 2020-06-08 2020-06-09
## 1 1 2 1 1 1 1
## 2020-06-10 2020-06-11 2020-06-12 2020-06-17 2020-06-20 2020-06-21 2020-06-22
## 1 1 1 1 2 2 1
## 2020-06-23 2020-06-25 2020-06-27 2020-06-28 2020-07-03 2020-07-04 2020-07-05
## 1 1 2 2 1 2 1
## 2020-07-08 2020-07-09 2020-07-10 2020-07-11 2020-07-12 2020-07-13 2020-07-15
## 1 1 1 2 2 1 1
## 2020-07-17 2020-07-18 2020-07-19 2020-07-20 2020-07-21 2020-07-25 2020-08-01
## 1 1 1 1 1 1 1
## 2020-08-02 2020-09-16 2020-09-20 2020-11-07 2020-11-15 2020-11-16 2020-12-02
## 1 1 1 1 1 1 1
## 2020-12-24 2020-12-25 2020-12-26 2020-12-27 2020-12-31 2021-01-01 2021-01-02
## 1 1 1 1 1 1 1
## 2021-01-03 2021-01-16 2021-01-17 2021-01-19 2021-01-22 2021-01-23 2021-01-24
## 1 1 1 1 1 1 1
## 2021-01-30 2021-01-31 2021-02-06 2021-02-07 2021-02-11 2021-02-13 2021-02-14
## 1 1 1 1 1 1 1
## 2021-02-16 2021-02-18 2021-02-20 2021-02-21 2021-02-23 2021-02-24 2021-02-25
## 1 1 1 1 1 1 1
## 2021-02-26 2021-02-27 2021-02-28 2021-03-04 2021-03-06 2021-03-07 2021-03-13
## 1 1 1 1 1 1 1
## 2021-03-14 2021-03-20 2021-03-21
## 1 1 1zero_cases_t=nordic[which(nordic$new_cases==0),]
#zero=zero_cases_t %>% group_by(date)%>% summarise(location=location)
#zeroBy reading this article McConnell points at the common mistakes for displaying trends. By using raw data, it is hard to distinguish the trends. Based on the situation above and this important reminder about correctly visualizing time series, I updated the plots using smoothed data.
We can now appreciate much better that, in Denmark, so far, there have been two similar picks in the number of cases in two different periods.
Norway had a significant number of cases at the beginning of the pandemic, which was much more control over the rest of the year until November where the trend starts rising again but still less pronounced.
In the cases of Finland, the numbers are low but constantly increasing (increasing trend). Sweden had a first short pick in June. In contrast, Iceland cases trend remain flat.
With more information about important holidays or dates in every countries culture, the details about what implied to break a residency rule for habitats, and much more, we could add labels and tags that complement this story.
Please find an updated version below
Normal version:
