07_time-dependent-graphs.Rmd

---
editor_options: 
  markdown: 
    wrap: 72
---

# Time-dependent graphs

**Learning objective:**

-   Discover different types of graphs for time series analysis

## Time Series

**Time-dependent graph** is a graph that structure changes with time.

**Time Series Analysis** is used to see how data behaves over a period
of time.

-   By using time series analysis, we can see seasonality change or
    cyclic behavior in the data

-   **Time Series Data** is observations indexed by time

## Economics Dataset

`economics` dataset comes from `ggplot2` package.

-   It contains US monthly economic data collected from January 1967
    through January 2015.

## Simple Line Plot

Let's plot personal savings rate using simple line plot.

-   `x = date`, `y = psavert`

```{r}
# load library
library(ggplot2)

ggplot(economics, aes(x = date, y = psavert)) +
  geom_line() +
  labs(title = "Personal Savings Rate",
       x = "Date",
       y = "Personal Savings Rate")
```

## Simple Line Plot +

Let us modify the previous graph.

-   `scale_x_date()` function from `scales` package can be used to
    reformat dates.

-   `theme_minimal()` is a function and it removes all background
    elements and reduces the axis lines and ticks to a minimum.

## Modified Simple Line Plot

```{r warning=FALSE, message=FALSE}
# load library
library(ggplot2)
library(scales)

ggplot(economics, aes(x = date, y = psavert)) +
  geom_line(color = "indianred3", 
            size=1 ) +
  geom_smooth() +
  scale_x_date(date_breaks = '5 year', 
               labels = date_format("%b-%y")) +
  labs(title = "Personal Savings Rate",
       subtitle = "1967 to 2015",
       x = "",
       y = "Personal Savings Rate") +
  theme_minimal()
```

## scale\_\*\_date Function

`scale_x_date()` `scale_y_date()` from `scales` package

Main parameters: `date_breaks()` `labels = date_format()`

-   *date_breaks()* is used to set the positions of the tick marks on
    the x-axis of a plot

-   *date_breaks()* function takes one argument, which is the unit of
    time

For example, we can use *date_breaks()* like this

date_breaks("1 year")

date_breaks("6 months")

date_breaks("1 week")

date_breaks("10 days")

-   *date_format()* function takes a single argument, which is a string
    that specifies the desired format of the date

For example,

date_format("%Y-%m-%d"): YYYY-MM-DD

date_format("%Y"): YYYY

date_format("%B"): Full month name (January, February, etc.)

date_format("%b"): Abbreviated month name (Jan, Feb, etc.)

date_format("%d"): Two-digit day of the month (01, 02, ..., 30, 31)

date_format("%A"): Full weekday name (Sunday, Monday, Tuesday, etc.)

date_format("%a"): Abbreviated weekday name (Sun, Mon, Tue, etc.)

more info:

[Scale_Date](https://ggplot2.tidyverse.org/reference/scale_date.html)

[Date_Format](https://www.rdocumentation.org/packages/base/versions/3.6.2/topics/strptime)

## Dumbbell Charts

-   **Dumbbell chart** is used to display the change between two time
    points for several groups or observations.

-   `geom_dumbbell()` function from `ggalt` package can be used

-   **Wide Format** dataset is used in this type of chart


## Long and Wide Format

[Long and Wide Formats in
Data](https://towardsdatascience.com/long-and-wide-formats-in-data-explained-e48d7c9a06cb)

![Long Format
Dataset](images/Screenshot%202023-05-07%20at%2013.01.36.png)

![Wide Format
Dataset](images/Screenshot%202023-05-07%20at%2013.00.29.png)


## Gapminder Dataset

`gapminder` dataset comes from `gapminder` package.

It is a subset of the [original data set](http://gapminder.org)

-   It contains for each of 142 countries, it provides values for life
    expectancy, GDP per capita, and population, every five years, from
    1952 to 2007.

## Dumbbell Charts

Let's plot the change in life expectancy from 1952 to 2007 in the
Americas.

```{r warning=FALSE, message=FALSE}
# load library
library(ggalt)
library(tidyr)
library(dplyr)

# load data
data(gapminder, package = "gapminder")

# subset data
plotdata_long <- filter(gapminder,
                        continent == "Americas" &
                        year %in% c(1952, 2007)) %>%
  select(country, year, lifeExp)

# convert data to wide format
plotdata_wide <- spread(plotdata_long, year, lifeExp)
names(plotdata_wide) <- c("country", "y1952", "y2007")

# create dumbbell plot
ggplot(plotdata_wide, aes(y = country,
                          x = y1952,
                          xend = y2007)) +  
  geom_dumbbell()
```

## Dumbbell Charts +

Let's modify a previous graph to make it easier to read. In order to do
so,

-   Sort by the countries

-   Change the color and size of lines and points

-   Use `theme_minimal()` to minimize the theme

## Modified Dumbbell Charts

```{r}
ggplot(plotdata_wide, 
       aes(y = reorder(country, y1952),
           x = y1952,
           xend = y2007)) +  
  geom_dumbbell(size = 1.2,
                size_x = 3, 
                size_xend = 3,
                colour = "grey", 
                colour_x = "blue", 
                colour_xend = "red") +
  theme_minimal() + 
  labs(title = "Change in Life Expectancy",
       subtitle = "1952 to 2007",
       x = "Life Expectancy (years)",
       y = "")
```

## Slope Graphs

-   *Slope graph* is useful when there are several groups and several
    time points.

-   `newggslopegraph()` function from the `CGPfunctions` package

## newggslopegraph Function

`newggslopegraph()` function takaes parameters in order.

newggslopegraph(*data frame*, *time variable*, *numeric variable*,
*grouping variable*)

-   **time variable** must be a factor

-   **numeric variable** will be plotted on the y axis

## Slope Graphs

Let's plot life expectancy for six Central American countries in 1992,
1997, 2002, and 2007.

```{r}
# load library
library(CGPfunctions)

# Select Central American countries data 
# for 1992, 1997, 2002, and 2007

df <- gapminder %>%
  filter(year %in% c(1992, 1997, 2002, 2007) &
           country %in% c("Panama", "Costa Rica", 
                          "Nicaragua", "Honduras", 
                          "El Salvador", "Guatemala",
                          "Belize")) %>%
  mutate(year = factor(year),
         lifeExp = round(lifeExp)) 

# create slope graph

newggslopegraph(df, year, lifeExp, country) +
  labs(title = "Life Expectancy by Country", 
       subtitle = "Central America", 
       caption = "source: gapminder")
```

## Area Charts

-   **Area Charts** is basically a line graph, with a fill from the line
    to the x-axis.
    
-   `geom_area()` is a function used to create an area plot. It is
    similar to a line plot, but instead of connecting the data points
    with a line, it fills in the area under the line.

```{r}
# basic area chart

ggplot(economics, aes(x = date, y = psavert)) +
  geom_area(fill="lightblue", color="black") +
  labs(title = "Personal Savings Rate",
       x = "Date",
       y = "Personal Savings Rate")
```

## Stacked Area Chart

-   **Stacked area chart** can be used to show differences between
    groups over time.

-   When interest is on both group change over time and overall change
    over time

## Uspopage Dataset

`uspopage` dataset from the `gcookbook` package

-   It contains age distribution of population in the United States,
    1900-2002

-   The values are estimated (not counted) by the U.S. Census

## Stacked Area Chart

Let's plot the age distribution of the US population from 1900 and 2002.

```{r}
# stacked area chart

# load data
data(uspopage, package = "gcookbook")

ggplot(uspopage, aes(x = Year,
                     y = Thousands, 
                     fill = AgeGroup)) +
  geom_area() +
  labs(title = "US Population by age",
       x = "Year",
       y = "Population in Thousands")
```

## Scientific Notation

How to change scientific notation such as $3e+05$ on the axis.

-   We can use `ggplot2` to change the scale.

1.  divide the Thousands variable by 1000

2.  report its unit as Millions

## Stacked Area Chart +

-   `fct_rev()` function from the `forcats` package allows us to reverse
    the levels of the specified variable

## Modified Stacked Area Chart

```{r}
# load data
data(uspopage, package = "gcookbook")

ggplot(uspopage, aes(x = Year,
                     y = Thousands/1000, 
                     fill = forcats::fct_rev(AgeGroup))) +
  geom_area(color = "black") +
  labs(title = "US Population by age",
       subtitle = "1900 to 2002",
       caption = "source: U.S. Census Bureau, 2003, HS-3",
       x = "Year",
       y = "Population in Millions",
       fill = "Age Group") +
  scale_fill_brewer(palette = "Set2") +
  theme_minimal()
```

## Meeting Videos {.unnumbered}

### Cohort 1 {.unnumbered}

`r knitr::include_url("https://www.youtube.com/embed/OtgqUi7kgzM")`

<details>

<summary>Meeting chat log</summary>

```         
00:11:17	Lydia Gibson:	Nice!
00:11:40	Oluwafemi Oyedele:	Reacted to "Nice!" with 👌
00:15:11	Lydia Gibson:	Cool. So the lowercase abbreviates whether its the month or day
00:18:19	Ken Vu:	Its good
00:18:20	Ken Vu:	:)
00:26:24	Lydia Gibson:	That’s beautiful 😍
00:26:25	Ken Vu:	That's pretty neat.  Looks very professional
00:31:06	Lydia Gibson:	Nice!
00:31:43	Ken Vu:	Colorful 🙂 Kind of lick a trendline version of stacked bar plots
00:31:46	Ken Vu:	*like
00:37:36	Ken Vu:	👏
00:37:47	Ken Vu:	Thank you, Kotomi!  You gave an awesome presentation
```

</details>