r-coding.Rmd

---
title: "Coding Notes"
description: |
  Coding Notes and Resources
author: 
  - name: Alex Stephenson
output:
  distill::distill_article:
    toc: true
    toc_depth: 3
site: distill::distill_website
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
options(scipen = 999)
```

```{r, message = F, warning = F}
library(broom)
library(data.table)
library(dplyr)
library(fixest)
library(ggplot2)
library(marginaleffects)
library(purrr)
```

## Basic R Operations 

There are a voluminous amount of resources on R online. Here is a very minimal crash course. 

The most important underlying concept in R is that (almost) everything is an object. 

### Data Types 

R has six basic data types. The four most common types we work with are vectors, lists, matrices, and data frames. 

#### Vectors 

```{r}
## Three ways to create a vector
v = c(1,2,3,4,5)
v2 = seq(from=1,to=5, by = 1)

## Since the sequence is counting with no breaks we can also do this
v3 = 1:5
```

Vectors can also be characters (strings in other languages) 

```{r}
w = c("A", "B", "C")
w2 = c("1","2","3")

## We can convert to other types by the as.* series 
w3 = as.character(v)
w4 = as.numeric(w2)
```


#### Lists 

```{r}
l = list(v, w)
l
```

#### Matrices 

```{r}
m = matrix(seq(1,16,1), nrow = 4, byrow = T)
m
```

For much more about linear algebra in R, consult the [Matrix Algebra in R](matrix.html) section. 

#### Data Frames 

```{r}
df = data.frame(
  var1 = letters[1:5],
  var2 = c(1,2,3,4,5)
)
df
```

### Subsetting 

### Control Operations 

### Functions

## Graphing in R 

### The Basics 

R has default plotting, but for the purpose of these notes I exclusively use `ggplot2`. A main distinction between base plotting and `ggplot2` is that `ggplot2` requires a data frame and presumes that your data is tidy.


```{r, echo = TRUE, message = F, warning = F}
## Using the built-in dataset diamonds
df = diamonds

## Data
df |>
  ## Mapping
  ggplot(aes(x = carat, y = price)) +
  ## geom
  geom_point() +
  ## we can have multiple geoms 
  geom_smooth(method = "lm", se = F)+
  ## Labels
  labs(x = "Carat",
       y = "Price",
       title = "Example Chart")+
  ## theme 
  theme_minimal()
```


For more extensive discussion of graphing capabilities in R, check out [Data Visualization: A Practical Introduction by Healy](https://socviz.co/). 

### Useful Plots for Political Science 

#### Coefficient Plot 

First let's make a model using the gapminder dataset. 

```{r}
## Make a model with robust standard errors using the gapminder
## dataset
m1 = fixest::feols(lifeExp ~ gdpPercap + pop + continent, data = gapminder::gapminder, vcov = "HC1") |>
  tidy(conf.int = TRUE)

head(m1)
```


```{r}
m1 |>
  ggplot(aes(x = term, y = estimate, 
             ymin = conf.low, 
             ymax = conf.high))+
  geom_point() +
  geom_pointrange() + 
  coord_flip() +
  theme_minimal() +
  labs(x = "",
       y = "Coefficient Estimates",
       title = "Coefficient Plot")
```

### Added Variable Plots 


### Marginal Effects Plots 

```{r}
marginModel = fixest::feglm(lifeExp ~ gdpPercap + pop + continent, data = gapminder::gapminder, vcov = "HC1") |>
  marginaleffects::slopes(by = TRUE)


```

## Simulations in R 

### The Birthday Problem