COVID-19 Global Analysis Dashboard

---
title: "COVID-19 Global Analysis Dashboard"
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    vertical_layout: fill
    theme: cosmo
    social: menu
    source_code: embed
---

```{r setup, include=FALSE}
# Load required libraries
library(flexdashboard)
library(tidyverse)
library(plotly)
library(DT)
library(lubridate)
library(scales)
library(viridis)

# Set options
knitr::opts_chunk$set(echo = FALSE, warning = FALSE, message = FALSE)
```

```{r load-data}
# Load and process data
covid <- read.csv("time_series_covid19_confirmed_global.csv")

# Transform to long format
covid_long <- covid %>%
  pivot_longer(
    cols = 5:ncol(covid),
    names_to = "Date",
    values_to = "Cases"
  ) %>%
  mutate(
    Date = str_replace(Date, "^X", ""),
    Date = mdy(Date)
  )

# Aggregate by country
covid_country <- covid_long %>%
  group_by(Country.Region, Date) %>%
  summarise(Total_Cases = sum(Cases, na.rm = TRUE), .groups = "drop") %>%
  arrange(Country.Region, Date) %>%
  group_by(Country.Region) %>%
  mutate(
    Daily_New_Cases = Total_Cases - lag(Total_Cases, default = 0),
    Daily_New_Cases = pmax(Daily_New_Cases, 0)
  ) %>%
  ungroup()

# Global totals
covid_global <- covid_country %>%
  group_by(Date) %>%
  summarise(
    Global_Total_Cases = sum(Total_Cases),
    Global_Daily_Cases = sum(Daily_New_Cases),
    .groups = "drop"
  )

# Get latest date and statistics
latest_date <- max(covid_global$Date)
latest_global <- covid_global %>% filter(Date == latest_date)

# Top 10 countries
top10_latest <- covid_country %>%
  filter(Date == latest_date) %>%
  arrange(desc(Total_Cases)) %>%
  slice(1:10)

top10_countries_names <- top10_latest$Country.Region

# Filter data for top 10 countries over time
top10_trends <- covid_country %>%
  filter(Country.Region %in% top10_countries_names)
```

# Overview {data-icon="fa-globe"}

## Column {data-width="350"}

### Global Total Cases

```{r}
valueBox(
  comma(latest_global$Global_Total_Cases),
  icon = "fa-users",
  color = "#d9534f"
)
```

### Daily New Cases

```{r}
valueBox(
  comma(latest_global$Global_Daily_Cases),
  icon = "fa-chart-line",
  color = "#f0ad4e"
)
```

### Countries Affected

```{r}
n_countries <- covid_country %>%
  filter(Date == latest_date, Total_Cases > 0) %>%
  nrow()

valueBox(
  n_countries,
  icon = "fa-flag",
  color = "#5bc0de"
)
```

### Data Coverage Period

```{r}
date_range <- paste(
  format(min(covid_global$Date), "%b %d, %Y"),
  "to",
  format(max(covid_global$Date), "%b %d, %Y")
)

valueBox(
  date_range,
  icon = "fa-calendar",
  color = "#5cb85c"
)
```

## Column {data-width="650"}

### Global Cumulative Cases Over Time

```{r}
p_global <- ggplot(covid_global, aes(x = Date, y = Global_Total_Cases)) +
  geom_line(color = "#d9534f", size = 1) +
  scale_y_continuous(labels = comma) +
  labs(
    x = "Date",
    y = "Total Confirmed Cases",
    title = NULL
  ) +
  theme_minimal(base_size = 12)

ggplotly(p_global) %>%
  layout(hovermode = "x unified")
```

### Top 10 Countries by Total Cases

```{r}
p_top10 <- ggplot(top10_latest, 
                  aes(x = reorder(Country.Region, Total_Cases), 
                      y = Total_Cases,
                      fill = Total_Cases)) +
  geom_col() +
  scale_y_continuous(labels = comma) +
  scale_fill_viridis_c(option = "plasma", labels = comma) +
  coord_flip() +
  labs(
    x = NULL,
    y = "Total Confirmed Cases",
    title = NULL
  ) +
  theme_minimal(base_size = 12) +
  theme(legend.position = "none")

ggplotly(p_top10)
```

# Country Trends {data-icon="fa-chart-line"}

## Column {data-width="600"}

### Time Series: Top 10 Countries

```{r}
p_trends <- ggplot(top10_trends, 
                   aes(x = Date, y = Total_Cases, color = Country.Region)) +
  geom_line(size = 0.8) +
  scale_y_continuous(labels = comma) +
  scale_color_viridis_d(option = "turbo") +
  labs(
    x = "Date",
    y = "Total Confirmed Cases",
    color = "Country",
    title = NULL
  ) +
  theme_minimal(base_size = 12) +
  theme(legend.position = "right")

ggplotly(p_trends) %>%
  layout(hovermode = "x unified", legend = list(orientation = "v"))
```

## Column {data-width="400"}

### Daily New Cases: Top 10 Countries

```{r}
p_daily <- ggplot(top10_trends, 
                  aes(x = Date, y = Daily_New_Cases, color = Country.Region)) +
  geom_line(size = 0.6, alpha = 0.7) +
  scale_y_continuous(labels = comma) +
  scale_color_viridis_d(option = "turbo") +
  labs(
    x = "Date",
    y = "Daily New Cases",
    color = "Country",
    title = NULL
  ) +
  theme_minimal(base_size = 11) +
  theme(legend.position = "right")

ggplotly(p_daily) %>%
  layout(hovermode = "x unified")
```

### Peak Daily Cases by Country

```{r}
peak_daily <- top10_trends %>%
  group_by(Country.Region) %>%
  summarise(
    Max_Daily_Cases = max(Daily_New_Cases, na.rm = TRUE),
    Peak_Date = Date[which.max(Daily_New_Cases)]
  ) %>%
  arrange(desc(Max_Daily_Cases))

datatable(
  peak_daily,
  colnames = c("Country", "Peak Daily Cases", "Peak Date"),
  options = list(
    pageLength = 10,
    dom = 't',
    ordering = TRUE
  ),
  rownames = FALSE
) %>%
  formatCurrency("Max_Daily_Cases", currency = "", digits = 0) %>%
  formatDate("Peak_Date", method = "toLocaleDateString")
```

# Comparative Analysis {data-icon="fa-balance-scale"}

## Column {data-width="500"}

### Heatmap: Cases Intensity Over Time (Top 10 Countries)

```{r}
# Prepare data for heatmap - sample every 7 days to reduce size
heatmap_data <- top10_trends %>%
  filter(wday(Date) == 1) %>%  # Keep only Sundays
  mutate(
    Month_Year = format(Date, "%Y-%m"),
    Cases_Log = log10(Total_Cases + 1)  # Log scale for better visualization
  )

p_heatmap <- ggplot(heatmap_data, 
                    aes(x = Date, y = Country.Region, fill = Cases_Log)) +
  geom_tile() +
  scale_fill_viridis_c(
    option = "magma",
    name = "Cases (log10)",
    labels = function(x) comma(10^x - 1)
  ) +
  labs(
    x = "Date",
    y = NULL,
    title = NULL
  ) +
  theme_minimal(base_size = 11) +
  theme(
    axis.text.x = element_text(angle = 45, hjust = 1),
    panel.grid = element_blank()
  )

ggplotly(p_heatmap)
```

## Column {data-width="500"}

### Growth Rate Comparison

```{r}
# Calculate growth rate (% change over 7 days)
growth_data <- covid_country %>%
  filter(Country.Region %in% top10_countries_names) %>%
  arrange(Country.Region, Date) %>%
  group_by(Country.Region) %>%
  mutate(
    Cases_7d_ago = lag(Total_Cases, 7),
    Growth_Rate = ((Total_Cases - Cases_7d_ago) / Cases_7d_ago) * 100
  ) %>%
  filter(Date >= latest_date - days(90)) %>%  # Last 90 days
  ungroup()

p_growth <- ggplot(growth_data, 
                   aes(x = Date, y = Growth_Rate, color = Country.Region)) +
  geom_line(size = 0.7, alpha = 0.8) +
  geom_hline(yintercept = 0, linetype = "dashed", color = "gray50") +
  scale_color_viridis_d(option = "turbo") +
  labs(
    x = "Date",
    y = "7-Day Growth Rate (%)",
    color = "Country",
    title = NULL
  ) +
  theme_minimal(base_size = 11)

ggplotly(p_growth) %>%
  layout(hovermode = "x unified")
```

### Latest Statistics: Top 10 Countries

```{r}
latest_stats <- covid_country %>%
  filter(Date == latest_date) %>%
  arrange(desc(Total_Cases)) %>%
  slice(1:10) %>%
  select(Country.Region, Total_Cases, Daily_New_Cases)

datatable(
  latest_stats,
  colnames = c("Country", "Total Cases", "Daily New Cases"),
  options = list(
    pageLength = 10,
    dom = 't',
    ordering = FALSE
  ),
  rownames = FALSE
) %>%
  formatCurrency(c("Total_Cases", "Daily_New_Cases"), 
                 currency = "", 
                 digits = 0)
```

# Data Table {data-icon="fa-table"}

### Complete Dataset (Latest Date)

```{r}
# Full data table for latest date
full_data <- covid_country %>%
  filter(Date == latest_date) %>%
  arrange(desc(Total_Cases)) %>%
  select(Country.Region, Total_Cases, Daily_New_Cases) %>%
  mutate(Rank = row_number())

datatable(
  full_data,
  colnames = c("Country/Region", "Total Cases", "Daily New Cases", "Rank"),
  options = list(
    pageLength = 25,
    order = list(list(3, 'asc')),
    searchHighlight = TRUE,
    autoWidth = TRUE
  ),
  filter = 'top',
  rownames = FALSE
) %>%
  formatCurrency(c("Total_Cases", "Daily_New_Cases"), 
                 currency = "", 
                 digits = 0)
```

# About {data-icon="fa-info-circle"}

## Column

### About This Dashboard

**COVID-19 Global Analysis Dashboard**

This interactive dashboard provides comprehensive visualization and analysis of COVID-19 confirmed cases globally.

**Data Source:** - Johns Hopkins University CSSE COVID-19 Data - Time Series Data: `time_series_covid19_confirmed_global.csv` - Repository: [GitHub](https://github.com/CSSEGISandData/COVID-19)

**Key Features:**

1.  **Overview Tab**: Global statistics and key metrics
2.  **Country Trends**: Time series analysis for top 10 affected countries
3.  **Comparative Analysis**: Heatmaps and growth rate comparisons
4.  **Data Table**: Searchable and sortable dataset

**Visualizations Include:**

-   Line charts showing cumulative and daily trends
-   Bar charts comparing countries
-   Heatmaps displaying intensity over time
-   Interactive tables with filtering capabilities

**Technical Stack:**

-   R version 4.4.2
-   Packages: `flexdashboard`, `tidyverse`, `plotly`, `DT`, `lubridate`

**Project Context:**

Data Visualization project to demonstrate:

-   Data collection and preprocessing skills
-   Exploratory data analysis techniques
-   Interactive dashboard design principles
-   Effective visual communication of complex data

**Data Coverage:**

-   Start Date: January 22, 2020
-   End Date: March 9, 2023
-   Geographic Coverage: Global (290+ regions/countries)

**Interactivity Features:**

-   Hover tooltips for detailed information
-   Zoom and pan capabilities on charts
-   Searchable and sortable data tables
-   Responsive design for various screen sizes

**Notes:**

-   Data reflects confirmed cases only (does not include deaths or recoveries)
-   Some countries include province/state subdivisions
-   Daily new cases calculated as day-over-day differences
-   Negative values (due to data corrections) are set to zero