Get_Truth_Ranks.m

%% OxCGRT

countries = unique(OxCGRT(:,1)); % Alphebetizes countries
n = size(countries,1);

% Get the number of entries by date
lowbound = find(OxCGRT.('Date')==20200123, 1);
upbound = find(OxCGRT.('Date')==20201120, 1); % Changed from 11/30 to 11/20

% Tracks the totals over all of the categories at each date
totals_OxCGRT = zeros(height(countries),upbound);

for i = 1:height(countries)
    % For each country
    c = countries.CountryName(i);
    country_data = OxCGRT(strcmp(OxCGRT.CountryName,c),:);
    
    % Get the columns for the first and last date for the country
    loc1 = find(country_data.('Date')==20200123, 1);
    loc2 = find(country_data.('Date')==20201120, 1); % Changed from 11/30 to 11/20
    
    tot = sum(country_data{loc1:loc2,[7:2:21,22,24,29:30,33]},2); % May remove 24
    
    totals_OxCGRT(i,lowbound:upbound) = tot;
    
end

totals_OxCGRT = table(countries.CountryName,totals_OxCGRT);

%% OneWorld

% Daily case rate zones classified by rate = (# new cases)/(100,000 residents)
% --> Examine (# new cases)/(total population)

countries = unique(OurWorld(:,3));

% Get the number of entries by date
lowbound = find(OurWorld.('date')==datetime([2020,01,23]), 1);
upbound = find(OurWorld.('date')==datetime([2020,11,20]), 1);

% Tracks the totals over all of the categories at each date
totals_OurWorld = zeros(height(countries),upbound);

for i = 1:height(countries)
    % For each country
    c = countries.location(i);
    country_data = OurWorld(strcmp(OurWorld.location,c),:);
    
    % Get the columns for the first and last date for the country
    loc1 = find(OurWorld.('date')==datetime([2020,01,23]), 1);
    loc2 = find(OurWorld.('date')==datetime([2020,11,20]), 1);
    
    rate = country_data{loc1:loc2,12}; % Column 12: new cases per million
    
    totals_OurWorld(i,lowbound:upbound) = rate;
    
end

totals_OurWorld = table(countries.location,totals_OurWorld);


%% Determine 'safe' and 'unsafe' countries

% Top (cutoff)% = 'safe'; Bottom (100-cutoff)% = 'unsafe'

% Examine number of new cases in the last month for each country
[~,idx] = sort(sum(totals_OurWorld.Var2(:,end-30:end),2));

% Select cutoff percentage:
b=[safe;unsafe];
b=b(:,2);
figure,plot(1:(length(b)),b)
hold on, xlabel('Country Rank in New Cases per Million')
ylabel('New Cases per Million (Averaged Over Last Month)')
title('New Cases per Million vs. Country Rank in New Cases per Million')

figure,loglog(1:(length(b)),b)
hold on, xlabel('Log of Country Rank in New Cases per Million')
ylabel('Log of New Cases per Million (Averaged Over Last Month)')
title('New Cases per Million vs. Country Rank in New Cases per Million (Log Plot)')



cutoff = 80/n; % Percentage cutoff: 100/n = 0.6 = 60% (due to above plot)

safe_countries = countries.location(idx(1:floor(n*cutoff)));
unsafe_countries = countries.location(idx(floor(n*cutoff)+1:end));

% [Government policy total; case total]
safe = [sum(totals_OxCGRT.Var2(idx(1:floor(n*cutoff)),end-30:end),2)/31,...
        sum(totals_OurWorld.Var2(idx(1:floor(n*cutoff)),end-30:end),2)/31];
unsafe = [sum(totals_OxCGRT.Var2(idx(floor(n*cutoff)+1:end),end-30:end),2)/31,...
        sum(totals_OurWorld.Var2(idx(floor(n*cutoff)+1:end),end-30:end),2)/31];

% Get the resulting table of safe and unsafe countries with the associated
% average of new cases per million people
safe_tab = table(safe_countries, safe(:,2),'VariableNames',{'Country','Avg. New Cases Per Million'});
unsafe_tab = table(unsafe_countries, unsafe(:,2),'VariableNames',{'Country','Avg. New Cases Per Million'});

% Normalize data
T = normalize([safe;unsafe],1);
safe_normalized = T(1:floor(n*cutoff),:);
unsafe_normalized = T(floor(n*cutoff)+1:end,:);