Team 5: The Impact of French Public Policies on Covid-19

Corentin Louison , Jérémy Grimault , Pierre Quemeneur
05-25-2020

Table of Contents

Pedagogical Material

As part of the Hacking Health Covid-19, the SKEMA Global Lab in AI provided to SKEMA’ students a fully developped data science environment to realize their project. See [here].

For this specific module, this team used these following courses:

Project Presentation

The main objective of the project is to allow the user to measure the impact of policydecisions taken to contain the expansion of Covid-19.Data We decided to work from the Oxford Covid tracker that we want to augment. We used their index and their updated database. Here is the link to the technical documentation on the Oxford project

Technical Process


library(dplyr)
library(readr)
library(lubridate)
library(ggplot2)
library(tidyr)
library(knitr)
library(kableExtra)

We based our code on this page: https://www.bsg.ox.ac.uk/research/research-projects/coronavirus-government-response-tracker

The link to the data is an extract from our Github, it would be better if we collect the daily actualized data from oxford: https://github.com/OxCGRT/covid-policy-tracker


data <- read_csv(file = "https://raw.githubusercontent.com/warint/covid-19/master/team5/data.csv")

You can find the explanation of the variables in the dataset [here]


dataCovid <- data %>%
  mutate (Date = ymd(Date)) %>% 
  select(-CountryCode, -`S8_Fiscal measures`,-`S9_Monetary measures`,
         -`S11_Investment in Vaccines`,-`S10_Emergency investment in health care`, -X27)

The next step was to define our thresholds for each strategy used by governments.


dataCovid <- mutate(dataCovid, 
                    School_closing = case_when(`S1_School closing` == 2 & 
                                                 S1_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Workplace_closing = case_when(`S2_Workplace closing` == 2 & 
                                                    S2_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Cancel_Public_Events = case_when(`S3_Cancel public events` == 2 & 
                                                       S3_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Close_Public_Transport = case_when(`S4_Close public transport` == 2 & 
                                                         S4_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Public_Information_Campaigns = case_when(`S5_Public information campaigns` == 1 & 
                                                               S5_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Restrictions_On_Internal_Movement = case_when(`S6_Restrictions on internal movement` == 2 & 
                                                                    S6_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    International_travel_controls = case_when(`S7_International travel controls` >= 2 ~ 1,TRUE ~ 0),
                    Limited_Testing_Policy = case_when(`S12_Testing framework` == 1 | 
                                                         `S12_Testing framework` == 2 ~ 1,TRUE ~ 0), 
                    Open_Testing_Policy = case_when(`S12_Testing framework` == 3 ~ 1,TRUE ~ 0),
                    Contact_tracing = case_when(`S13_Contact tracing` >= 1 ~ 1,TRUE ~ 0))

Because we wanted to analyse France response to covid-19, we had to filter to only keep France’s data. On this graph, we can see the evolution of French confirmed cases over time.


#Selection of a country, France,  among 193 countries
dataCovidFrance <- dataCovid %>%
  filter (CountryName == "France")

#Visualisation of French Confirmed Cases over Time
ggplot(dataCovidFrance,aes(x=Date, y=ConfirmedCases, col=StringencyIndex))+
  geom_line()+
  ggtitle("French Confirmed Cases over Time")+
  scale_y_log10() #we put y at a log10 scale

After that, we had to select the right public policies based on our threshold.


#Pivot the Covid France dataset to gather the observations regarding the public policies
dataCovidFrance_td <- dataCovidFrance %>% 
  select(CountryName, Date, ConfirmedCases, ConfirmedDeaths, StringencyIndex, School_closing,
         Workplace_closing,
         Cancel_Public_Events ,
         Close_Public_Transport ,
         Public_Information_Campaigns,
         Restrictions_On_Internal_Movement,
         International_travel_controls ,
         Limited_Testing_Policy , 
         Open_Testing_Policy,
         Contact_tracing) %>%
  pivot_longer(-c(CountryName, Date, ConfirmedCases, ConfirmedDeaths, StringencyIndex), 
               names_to = "policy", values_to = "value")

#select the observation when the policies are effective based on our threshold
test4 <- dataCovidFrance_td %>% 
  filter(value == 1) %>%
  select(CountryName, Date, policy, value)

#dataframe with the dates of the implementation of each policies
start <- test4 %>%
  group_by(policy, value) %>%
  slice(1)

#dataframe with the dates of the end of each policies
end <- test4 %>%
  group_by(policy, value) %>%
  arrange(value) %>%  
  slice(n())

#dataset gathering implementation and end dates for each policies
startend <- left_join(start, end, by = c("CountryName", "policy", "value"))

startend <- startend %>% 
  rename(
    start = Date.x,
    end = Date.y
  )

Finally we created a data visualization of the implementation of the French public policies regarding the number of confirmed case over time


library(ggrepel)

 ggplot(dataCovidFrance_td, aes(x=Date, y=ConfirmedCases)) + 
  geom_line(
    size = 1) +
  geom_vline(
    data = startend,
    aes(xintercept = jitter(as.numeric(start)), col=policy),
    alpha = 0.5,
    size = 1)+
  geom_label_repel(
    aes(x = start , y = 50 , label = policy, fill = policy),
    data = startend,
    position = "jitter",
    colour = "white", 
    fontface = "bold",
    segment.color = "grey50", force = 2)+
  ggtitle("French Public Policies and Confirmed Cases over Time")+
  scale_y_log10("Confirmed Cases") + 
  scale_x_date('Time', date_labels = "%e %b", date_breaks = "1 week")+
  theme_classic()+
  theme(plot.title=element_text(family='', face='bold', colour='black', size=12),
        axis.text.x = element_text(margin = margin(t = .3, unit = "cm"), angle = 90),
        legend.position = "none")

Here is the same graphic, but animated.


library(gganimate)
theme_set(theme_bw())
anim <- p + transition_reveal(Date)

# What still need to be done :
# 1- Add a nice animation 
# 2- Plug to daily updated database

##if we want to compare Public policies & ConfirmedDeaths
  geom_line(
    data = dataCovidFrance_td,
    aes(x=Date, y=ConfirmedDeaths),
    colour = "red")

mapping: x = ~Date, y = ~ConfirmedDeaths 
geom_line: na.rm = FALSE, orientation = NA
stat_identity: na.rm = FALSE
position_identity

tl;dr


library(dplyr)
library(readr)
library(lubridate)
library(ggplot2)
library(tidyr)
library(knitr)
library(kableExtra)

data <- read_csv(file = "https://raw.githubusercontent.com/warint/covid-19/master/team5/data.csv")

dataCovid <- data %>%
  mutate (Date = ymd(Date)) %>% 
  select(-CountryCode, -`S8_Fiscal measures`,-`S9_Monetary measures`,-`S11_Investment in Vaccines`,
         -`S10_Emergency investment in health care`, -X27)

dataCovid <- mutate(dataCovid, 
                    School_closing = case_when(`S1_School closing` == 2 & 
                                                 S1_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Workplace_closing = case_when(`S2_Workplace closing` == 2 & 
                                                    S2_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Cancel_Public_Events = case_when(`S3_Cancel public events` == 2 & 
                                                       S3_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Close_Public_Transport = case_when(`S4_Close public transport` == 2 & 
                                                         S4_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Public_Information_Campaigns = case_when(`S5_Public information campaigns` == 1 & 
                                                               S5_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    Restrictions_On_Internal_Movement = case_when(`S6_Restrictions on internal movement` == 2 & 
                                                                    S6_IsGeneral == 1 ~ 1,TRUE ~ 0),
                    International_travel_controls = case_when(`S7_International travel controls` >= 2 ~ 1,TRUE ~ 0),
                    Limited_Testing_Policy = case_when(`S12_Testing framework` == 1 | 
                                                         `S12_Testing framework` == 2 ~ 1,TRUE ~ 0), 
                    Open_Testing_Policy = case_when(`S12_Testing framework` == 3 ~ 1,TRUE ~ 0),
                    Contact_tracing = case_when(`S13_Contact tracing` >= 1 ~ 1,TRUE ~ 0))

#Selection of a country, France,  among 193 countries
dataCovidFrance <- dataCovid %>%
  filter (CountryName == "France")

#Visualisation of French Confirmed Cases over Time
ggplot(dataCovidFrance,aes(x=Date, y=ConfirmedCases, col=StringencyIndex))+
  geom_line()+
  ggtitle("French Confirmed Cases over Time")+
  scale_y_log10() #we put y at a log10 scale

#Pivot the Covid France dataset to gather the observations regarding the public policies
dataCovidFrance_td <- dataCovidFrance %>% 
  select(CountryName, Date, ConfirmedCases, ConfirmedDeaths, StringencyIndex, School_closing,
         Workplace_closing,
         Cancel_Public_Events ,
         Close_Public_Transport ,
         Public_Information_Campaigns,
         Restrictions_On_Internal_Movement,
         International_travel_controls ,
         Limited_Testing_Policy , 
         Open_Testing_Policy,
         Contact_tracing) %>%
  pivot_longer(-c(CountryName,Date, ConfirmedCases,ConfirmedDeaths, StringencyIndex), 
               names_to = "policy", values_to = "value")

#select the observation when the policies are effective based on our threshold
test4 <- dataCovidFrance_td %>% 
  filter(value == 1) %>%
  select(CountryName, Date, policy, value)

#dataframe with the dates of the implementation of each policies
start <- test4 %>%
  group_by(policy, value) %>%
  slice(1)

#dataframe with the dates of the end of each policies
end <- test4 %>%
  group_by(policy, value) %>%
  arrange(value) %>%  
  slice(n())

#dataset gathering implementation and end dates for each policies
startend <- left_join(start, end, by = c("CountryName", "policy", "value"))

startend <- startend %>% 
  rename(
    start = Date.x,
    end = Date.y
  )

#Datavisualization
library(ggrepel)

 ggplot(dataCovidFrance_td, aes(x=Date, y=ConfirmedCases)) + 
  geom_line(
    size = 1) +
  geom_vline(
    data = startend,
    aes(xintercept = jitter(as.numeric(start)), col=policy),
    alpha = 0.5,
    size = 1)+
  geom_label_repel(
    aes(x = start , y = 50 , label = policy, fill = policy),
    data = startend,
    position = "jitter",
    colour = "white", 
    fontface = "bold",
    segment.color = "grey50", force = 2)+
  ggtitle("French Public Policies and Confirmed Cases over Time")+
  scale_y_log10("Confirmed Cases") + 
  scale_x_date('Time', date_labels = "%e %b", date_breaks = "1 week")+
  theme_classic()+
  theme(plot.title=element_text(family='', face='bold', colour='black', size=12),
        axis.text.x = element_text(margin = margin(t = .3, unit = "cm"), angle = 90),
        legend.position = "none")


library(gganimate)
theme_set(theme_bw())
p + transition_reveal(Date)


# What still need to be done :
# 1- Add a nice animation 
# 2- Plug to daily updated database



##if we want to compare Public policies & ConfirmedDeaths
  geom_line(
    data = dataCovidFrance_td,
    aes(x=Date, y=ConfirmedDeaths),
    colour = "red")

To go further with our pedagogical platform

Reuse

Text and figures are licensed under Creative Commons Attribution CC BY 4.0. The figures that have been reused from other sources don't fall under this license and can be recognized by a note in their caption: "Figure from ...".

© Copyright 2020 Nüance-R and the SKEMA Global Lab in AI

Software licensed under the MIT License.