Data Science and Statistical Computing

class: center, middle, inverse, title-slide

.title[
# Data Science and Statistical Computing
]
.subtitle[
## Practical Lecture 7<br>Dynamic Documents
]
.author[
### Dr Louis Aslett
]
.institute[
### Durham University
]
.date[
### 21 November 2024
]

---

## Dynamic documents?

**What?**

- Combines report writing/slides/etc *and* code *and* data source in one file.
- Change input data and the document will dynamically update.
- Single source for multiple output formats (HTML, PDF, Word)
- Single language for multiple output types (presentations, reports, books, papers, ...)

**Why?**

A major challenge for using data in industry, commercial and research settings is that reporting can become separated from the data itself, leading to problems in tracing how a particular graphic, statistic, etc was produced.

- Harms reproducibility
- Slows down making changes to graphics/analysis
- Updating with new data slow
- Error prone if process by which document generated is obfuscated

RMarkdown allows us to mix content and code in a *plain text* format (no WYSIWYG).

---

background-image: url(i/word_meme.jpg)
background-size: contain
class: center, bottom, inverse

---

class: inverse

## Getting the packages

``` r
install.packages("rmarkdown")
install.packages("tinytex")
```

`tinytex` includes a miniture LaTeX distribution so that you can compile to PDFs!

---

## RMarkdown

- Simple plain text format

- Separate presentational style from content

- Simplified formatting commands

- Embedded code

- Include LaTeX for mathematical content

![](i/rmarkdown.jpg)

---

## Document preamble

Section enclosed between two lines of `---` at the top is the "document preamble".

Specifies various options for the document, including:

- `title:` the document title
- `author:` (optional) author's name
- `date:` (optional) date for the document
- `output:` a tree like structure for setting options of different output formats (HTML/PDF/Word)

---

## Simple formatting

Styles:

- `*italic text*`
    - *italic text* 
- `**bold text**`
    - **bold text**
- `~~strikeout text~~`
    - ~~strikeout text~~

Sections:

- `# Section heading`
    - 1. Section heading
- `## Sub-section heading`
    - 1.1 Sub-section heading
- `### Sub-sub-section heading`
    - 1.1.1 Sub-sub-section heading

Misc:

- `[Dr Aslett's homepage](https://www.louisaslett.com/)`
    - [Dr Aslett's homepage](https://www.louisaslett.com/)

---

## Lists

.pull-left[
- First item
- Second item
    - First sub-item
    - Second sub-item
- Third item
]

.pull-right[

``` r
- First item
- Second item
    - First sub-item
    - Second sub-item
- Third item
```
]

.pull-left[

1. First item
1. Second item

i. First sub-item
    
    ii. Second sub-item

1. Third item

]

.pull-right[

``` r
1. First item
1. Second item
    i. First sub-item
    i. Second sub-item
1. Third item
```
]

**Note:** 4 spaces needed to create the indent

**Note:** Don't need to increment the numbering manually

---

## Including R code and output (inline)

R can be included inline in text very easily. eg:

``` r
`r 1+1`
```

will substitute the value 2 in that place.

To show the code in a code font, but not execute, don't include the "r". eg:

``` r
`1+1`
```

will make the text look like this `1+1`.

---

## Including R code and output (chunks) I

Use the triple-backtick fence, specifying the language to be R.

Prints both the code (nicely formatted) and the result of executing that code:

``` r
x <- rnorm(10, 3)
mean(x)
```

```
[1] 3.317083
```

---

## Including R code and output (chunks) II

Various options change the behaviour, including:

- `echo=FALSE` to suppress printing the code (just show the output)
- `eval=FALSE` to show the code, but not run it

Obviously using both together is ... largely pointless!

Plots can be included by simply calling the plotting command.

- `fig.width` and `fig.height` take numeric options to specify size
- `out.width` and `out.height` allow character options (eg "25%") to proportionately scale.

---

<code>
&#96;&#96;&#96;{r,echo=FALSE} <br>
hist(x, main="Histogram of 100 Normal(mean=3, sd=1)") <br>
&#96;&#96;&#96;
</code>

![](07_rmarkdown_files/figure-html/unnamed-chunk-9-1.png)

---

## Tables

``` html
| Col1 | Col2 | Col3 | Col4 | Col5 |
|------|------|-----:|------|:----:|
| a    | 1    | x    | `a`  | x    |
| b    | 2    | y    | `b`  | y    |
```

| Col1 | Col2 | Col3 | Col4 | Col5 |
|------|------|-----:|------|:----:|
| a    | 1    | x    | `a`  | x    |
| b    | 2    | y    | `b`  | y    |

See also the great package `gt` [https://gt.rstudio.com](https://gt.rstudio.com)

---

## Maths

You can write LaTeX inside a pair of dollar symbols, eg:

`$\alpha+\beta$` renders `$\alpha+\beta$`

You can use the display style with double dollar signs:

```
$$\bar{X}=\frac{1}{n}\sum_{i=1}^nX_i$$
```

`$$\bar{X}=\frac{1}{n}\sum_{i=1}^nX_i$$`

```
\begin{align*}
asd &= asd \\
&= asd
\end{align*}
```

`\begin{align*}
e^{i \pi} &= \cos \pi + i \sin \pi \\
&= -1
\end{align*}`

---

## Presentations

Exactly the same source can support generating presentations rather than documents!

Main difference: `##` triggers a new slide.

---

## Resources