Introduction

In this lesson we will learn how to create

Scientific reproducible reports, books, papers…
Slides
Statistical Dashboards

Before starting you have to make sure that LaTeX is installed in your computer:

Download Miktex (for Widows users) and Mactex (for Mac users)
Online and collaborative tools : Overleaf and Sharelatex

Producing Reports

Overleaf

Create your own account
For your first document choose one template and change it! (Here’s a demo)
We will show:
- How to add a figure and a table: Caption, position in the text, reference and labels?
- How to add Bibliography: .bib file, several options: bibtex code: Jabref (for windows users) and Bibdeck (for Mac users) Or Google scholar

`rmarkdown`:

Document
- HTML (for websites)
- PDF (in Latex)
- Word document
Presentation
- HTML: 2 types isoslides and slidy
- PDF: beamer presentations
- PowerPoint (see Example below)
From template
- Revealjs (Presentation)
- Tufte (HTML and PDF)
- Bioconductor (PDF)
- Flexdashboard

Flexdashboard (from Template)

YAML

 ---
title: "rbokeh iris dataset"
author: "Ryan Hafen"
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    social: menu
    source_code: embed
---

Loading packages that will be used

library(rbokeh)
library(flexdashboard)

Part 1
- Declare the column

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

+ Make your plot

figure(width = NULL, height = NULL) %%
  ly_points(Sepal.Length, Sepal.Width, data = iris, color = Species)

Part 2
- Declare 2nd column

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

Figure 2

figure(width = NULL, height = NULL, legend_location = "top_left") %%
  ly_quantile(Sepal.Length, group = Species, data = iris)

and

figure(width = NULL, height = NULL) %%
  ly_points(Sepal.Length, Sepal.Width, data = iris,
    color = Petal.Width)

Flexdashboard with Shiny app

YAML

---
title: "Iris K-Means Clustering"
output: 
  flexdashboard::flex_dashboard:
    orientation: columns
    social: menu
    source_code: embed
runtime: shiny
---

Loading packages

library(datasets)
data(iris)

Declaring the sidebar

Column {.sidebar}
-----------------------------------------------------------------------

Input part (UI)

selectInput('xcol', 'X Variable', names(iris))
    
selectInput('ycol', 'Y Variable', names(iris),
                selected=names(iris)[[2]])
    
numericInput('clusters', 'Cluster count', 3,
              min = 1, max = 9)

Declaring the Main panel

Column
-----------------------------------------------------------------------

Displaying results

palette(c("#E41A1C", "#377EB8", "#4DAF4A", "#984EA3",
  "#FF7F00", "#FFFF33", "#A65628", "#F781BF", "#999999"))

selectedData <- reactive({
  iris[, c(input$xcol, input$ycol)]
})

clusters <- reactive({
  kmeans(selectedData(), input$clusters)
})

renderPlot({
  par(mar = c(5.1, 4.1, 0, 1))
  plot(selectedData(),
       col = clusters()$cluster,
       pch = 20, cex = 3)
  points(clusters()$centers, pch = 4, cex = 4, lwd = 4)
})

Power Point with `R` (Example)

Installing packages:

> library(officer)
> library(magrittr)

Create an R object representing a PowerPoint document.

> my_pres <- read_pptx()

To add a slide in your presentation

> my_pres <- my_pres %>% 
+   add_slide(layout = "Title and Content", master = "Office Theme")

You can now generate you pptx file

> print(my_pres, target = "first_example.pptx")

We can check if the file is created

> file.exists("first_example.pptx") ## The answer should be TRUE

We can now manipulate my_pres object to create more slides, write text on the slides or add figures:

A slide can be added with the add_slide() function.

> my_pres <- read_pptx() %>% 
+   add_slide(layout = "Two Content", master = "Office Theme")

Our document will contain 1 slide

> length(my_pres)

We will add a text to the first slide and print in the pptx file

> my_pres <- my_pres %>% 
+   ph_with_text(type = "title", str = "A title", index = 1)
> 
> print(my_pres, target = "first_example.pptx")

We add date in the footnote

> my_pres<-my_pres %>%
+   ph_with_text(type = "dt", str = format(Sys.Date()), index = 1)
> 
> print(my_pres, target = "first_example.pptx")

We add a body text

> my_pres <- my_pres %>% 
+   ph_with_text(type = "body", str = "A first text", index = 1)
> 
> print(my_pres, target = "first_example.pptx")

we add an image

> img.file <- file.path( R.home("doc"), "html", "logo.jpg" )
> 
> my_pres <- my_pres %>% 
+   ph_with_img(type = "body", index = 1, src = img.file, height = 1.06, width = 1.39 )

We will add now a ggplot figure in a new file. We will first add a slide without a title and “two contents”

> my_pres<-my_pres%>%
+   add_slide(layout = "Two Content", master = "Office Theme")

We create the ggplot

> library(ggplot2)
> gg_plot <- ggplot(data = iris ) +
+   geom_point(mapping = aes(Sepal.Length, Petal.Length), size = 3) +
+   theme_bw()
> gg_plot

We will add on the left column of the slide

> if( capabilities(what = "png") )
+   my_pres <- ph_with_gg(my_pres, value = gg_plot, index=1 )

We can then add a text in the right column of the slide

> my_pres <- my_pres %>%
+   ph_with_text(type = "body", str = "Iris Data", index = 2)
> 
> ## Save the file
> print(my_pres, target = "first_example.pptx")

More ressources on `officer`

Plots in LaTeX format: `tikzDevice`

We want to change this figure into LaTeX code

>  library(ggpubr)
>  data("ToothGrowth")
>  ggboxplot(ToothGrowth, x = "dose", y = "len", width = 0.8)

library(tikzDevice)
tikz("fig.tex",standAlone=TRUE)
 ggboxplot(df, x = "dose", y = "len", width = 0.8)
Measuring dimensions of: \char77
Measuring dimensions of: \char100
Measuring dimensions of: \char111
....
 dev.off()
null device 
          1

Latex file is created in your work directory

% Created by tikzDevice version 0.10.1 on 2017-11-21 18:43:00
% !TEX encoding = UTF-8 Unicode
\documentclass[10pt]{article}
\usepackage{tikz}

\usepackage[active,tightpage,psfixbb]{preview}

\PreviewEnvironment{pgfpicture}

\setlength\PreviewBorder{0pt}
\begin{document}

\begin{tikzpicture}[x=1pt,y=1pt]
\definecolor{fillColor}{RGB}{255,255,255}
.... # truncated output.

`HTML` tables: `DT` package

Basic example

>  library(DT)
>  datatable(iris, options = list(
+  searching = FALSE,
+    pageLength = 5,
+    lengthMenu = c(5, 15, 35, 50)
+  ))

`DT` tables with options: color in one column

> datatable(iris, options = list(pageLength = 5)) %>%
+   formatStyle('Sepal.Length',  color = 'red', backgroundColor = 'orange', fontWeight = 'bold')

Adding buttons to download the table

> datatable(
+   iris, extensions = 'Buttons', options = list(
+     dom = 'Bfrtip',
+     buttons = c('copy', 'csv', 'excel', 'pdf', 'print')
+   )
+ )

Reordering columns

You can click and drag the table header to move a certain column to a different place

> datatable(iris[1:10,], extensions = 'ColReorder', options = list(colReorder = TRUE))

More ressources on `DT` package

HTML table with `sjPlot` package

> library(sjmisc)
> library(sjPlot)
> xCSS = list(css.table = "border: 2px solid;",
+                 css.tdata = "border: 1px solid;",
+                 css.firsttablecol = "color:#003399;
+                 font-weight:bold;")
> data(efc)
> tab_df(efc[1:5,], CSS=xCSS)

c12hour	e15relat	e16sex	e17age	e42dep	c82cop1	c83cop2	c84cop3	c85cop4	c86cop5	c87cop6	c88cop7	c89cop8	c90cop9	c160age	c161sex	c172code	c175empl	barthtot	neg_c_7	pos_v_4	quol_5	resttotn	tot_sc_e	n4pstu	nur_pst
16	2	2	83	3	3	2	2	2	1	1	2	3	3	56	2	2	1	75	12	12	14	0	4	0	NA
148	2	2	88	3	3	3	3	3	4	1	3	2	2	54	2	2	1	75	20	11	10	4	0	0	NA
70	1	2	82	3	2	2	1	4	1	1	1	4	3	80	1	1	0	35	11	13	7	0	1	2	2
168	1	2	67	4	4	1	3	1	1	1	1	2	4	69	1	2	0	0	10	15	12	2	0	3	3
168	2	2	84	4	3	2	1	2	2	2	1	4	4	47	2	2	0	25	12	15	19	2	1	2	2

Online LaTex and HTML tables

https://www.tablesgenerator.com

LaTeX and tables with `xtable` package

LaTeX

>  library(xtable)
> data(tli)
> head(tli)

grade sex disadvg ethnicty tlimth 1 6 M YES HISPANIC 43 2 7 M NO BLACK 88 3 5 F YES HISPANIC 34 4 3 M YES HISPANIC 65 5 8 M YES WHITE 75 6 5 M NO BLACK 74

> tli$grade<-tli$grade+rnorm(nrow(tli))
>  tli.table <- xtable(tli[1:3,],align=c("l","c","l","l","c","c"),
+                      digits=c(0,2,0,0,0,0),
+                      caption="My table",label="tab1")
> tli.table

% latex table generated in R 3.5.1 by xtable 1.8-3 package % Fri Nov 2 09:25:24 2018

HTML

> print(tli.table,type="html")

Models in LaTeX and HTML tables with `stargazer` package

Data Summary

>  library(stargazer)
>  data("attitude")
>  head(attitude,3)

rating complaints privileges learning raises critical advance 1 43 51 30 39 61 92 45 2 63 64 51 54 63 73 47 3 71 70 68 69 76 86 48

>  stargazer(attitude,flip=T)

% Table created by stargazer v.5.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu % Date and time: Fri, Nov 02, 2018 - 09:25:24

Display the data

>  stargazer(attitude[1:5,], summary=FALSE)

% Table created by stargazer v.5.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu % Date and time: Fri, Nov 02, 2018 - 09:25:24

Displaying Estimating and comparing Models

> library(stargazer)
> attitude$high.rating <- (attitude$rating > 70)
> linear.1 <- lm(rating ~ complaints + privileges + learning 
+                         + raises + critical, data=attitude)
> 
> linear.2 <- lm(rating ~ complaints + privileges + learning, data=attitude)
> probit.model <- glm(high.rating ~ learning + critical + advance, data=attitude,
+                     family = binomial(link = "probit"))
>  
> stargazer(linear.1, linear.2, probit.model, title="Regression Results")

% Table created by stargazer v.5.2 by Marek Hlavac, Harvard University. E-mail: hlavac at fas.harvard.edu % Date and time: Fri, Nov 02, 2018 - 09:25:24

In HTML format

> library(stargazer)
> attitude$high.rating <- (attitude$rating>70)
> linear.1 <- lm(rating ~ complaints + privileges + learning 
+                         + raises + critical, data=attitude)
> 
> linear.2 <- lm(rating ~ complaints + privileges + learning, data=attitude)
> probit.model <- glm(high.rating ~ learning + critical + advance, data=attitude,
+                     family = binomial(link = "probit"))
>  
> stargazer(linear.1, linear.2, probit.model,
+           type="html",
+           title="Regression Results",
+           single.row=TRUE,
+           ci=TRUE, ci.level=0.9, 
+           omit.stat=c("f", "ser"))

**Regression Results**

	Dependent variable:

	rating		high.rating
	OLS		probit
	(1)	(2)	(3)

complaints	0.692^*** (0.447, 0.937)	0.682^*** (0.470, 0.894)
privileges	-0.104 (-0.325, 0.118)	-0.103 (-0.316, 0.109)
learning	0.249 (-0.013, 0.512)	0.238^* (0.009, 0.467)	0.164^*** (0.077, 0.252)
raises	-0.033 (-0.366, 0.299)
critical	0.015 (-0.227, 0.258)		-0.001 (-0.073, 0.072)
advance			-0.062 (-0.131, 0.007)
Constant	11.011 (-8.240, 30.262)	11.258 (-0.779, 23.296)	-7.476^** (-13.349, -1.604)

Observations	30	30	30
R²	0.715	0.715
Adjusted R²	0.656	0.682
Log Likelihood			-9.087
Akaike Inf. Crit.			26.175

Note:	p<0.1; p<0.05; p<0.01

c12hour	e15relat	e16sex	e17age	e42dep	c82cop1	c83cop2	c84cop3	c85cop4	c86cop5	c87cop6	c88cop7	c89cop8	c90cop9	c160age	c161sex	c172code	c175empl	barthtot	neg_c_7	pos_v_4	quol_5	resttotn	tot_sc_e	n4pstu	nur_pst
16	2	2	83	3	3	2	2	2	1	1	2	3	3	56	2	2	1	75	12	12	14	0	4	0	NA
148	2	2	88	3	3	3	3	3	4	1	3	2	2	54	2	2	1	75	20	11	10	4	0	0	NA
70	1	2	82	3	2	2	1	4	1	1	1	4	3	80	1	1	0	35	11	13	7	0	1	2	2
168	1	2	67	4	4	1	3	1	1	1	1	2	4	69	1	2	0	0	10	15	12	2	0	3	3
168	2	2	84	4	3	2	1	2	2	2	1	4	4	47	2	2	0	25	12	15	19	2	1	2	2

c12hour	e15relat	e16sex	e17age	e42dep	c82cop1	c83cop2	c84cop3	c85cop4	c86cop5	c87cop6	c88cop7	c89cop8	c90cop9	c160age	c161sex	c172code	c175empl	barthtot	neg_c_7	pos_v_4	quol_5	resttotn	tot_sc_e	n4pstu	nur_pst
16	2	2	83	3	3	2	2	2	1	1	2	3	3	56	2	2	1	75	12	12	14	0	4	0	NA
148	2	2	88	3	3	3	3	3	4	1	3	2	2	54	2	2	1	75	20	11	10	4	0	0	NA
70	1	2	82	3	2	2	1	4	1	1	1	4	3	80	1	1	0	35	11	13	7	0	1	2	2
168	1	2	67	4	4	1	3	1	1	1	1	2	4	69	1	2	0	0	10	15	12	2	0	3	3
168	2	2	84	4	3	2	1	2	2	2	1	4	4	47	2	2	0	25	12	15	19	2	1	2	2

Introduction

Producing Reports

Overleaf

rmarkdown:

Flexdashboard (from Template)

Flexdashboard with Shiny app

Power Point with R (Example)

More ressources on officer

Plots in LaTeX format: tikzDevice

HTML tables: DT package

Basic example

DT tables with options: color in one column

Adding buttons to download the table

Reordering columns

More ressources on DT package

HTML table with sjPlot package

Online LaTex and HTML tables

LaTeX and tables with xtable package

LaTeX

HTML

Models in LaTeX and HTML tables with stargazer package

Data Summary

Display the data

Displaying Estimating and comparing Models

In HTML format

`rmarkdown`:

Power Point with `R` (Example)

More ressources on `officer`

Plots in LaTeX format: `tikzDevice`

`HTML` tables: `DT` package

`DT` tables with options: color in one column

More ressources on `DT` package

HTML table with `sjPlot` package

LaTeX and tables with `xtable` package

Models in LaTeX and HTML tables with `stargazer` package

c12hour	e15relat	e16sex	e17age	e42dep	c82cop1	c83cop2	c84cop3	c85cop4	c86cop5	c87cop6	c88cop7	c89cop8	c90cop9	c160age	c161sex	c172code	c175empl	barthtot	neg_c_7	pos_v_4	quol_5	resttotn	tot_sc_e	n4pstu	nur_pst
16	2	2	83	3	3	2	2	2	1	1	2	3	3	56	2	2	1	75	12	12	14	0	4	0	NA
148	2	2	88	3	3	3	3	3	4	1	3	2	2	54	2	2	1	75	20	11	10	4	0	0	NA
70	1	2	82	3	2	2	1	4	1	1	1	4	3	80	1	1	0	35	11	13	7	0	1	2	2
168	1	2	67	4	4	1	3	1	1	1	1	2	4	69	1	2	0	0	10	15	12	2	0	3	3
168	2	2	84	4	3	2	1	2	2	2	1	4	4	47	2	2	0	25	12	15	19	2	1	2	2