### Install eventstudyr Package

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Install the eventstudyr package from CRAN or the latest version from GitHub using devtools.

```r
install.packages("eventstudyr")
```

```r
install.packages("devtools")
devtools::install_github("JMSLab/eventstudyr")
```

--------------------------------

### Install eventstudyr from GitHub

Source: https://github.com/jmslab/eventstudyr/blob/main/README.md

Installs the latest version of the eventstudyr package directly from GitHub using the devtools package.

```r
install.packages("devtools")
devtools::install_github("JMSLab/eventstudyr")
```

--------------------------------

### Install eventstudyr from CRAN

Source: https://github.com/jmslab/eventstudyr/blob/main/README.md

Installs the eventstudyr package from the Comprehensive R Archive Network (CRAN).

```r
install.packages("eventstudyr")
```

--------------------------------

### Load and Utilize Sample Panel Data

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Access the example_data dataset to demonstrate event study workflows and model estimation.

```r
library(eventstudyr)

# Load and inspect sample data
data(example_data)
dim(example_data)
# [1] 2000   12

head(example_data)
# A tibble showing: id, t, y_base, y_jump_m, y_smooth_m, z, x_r, etc.

# View documentation
?example_data

# Key variables:
# - id: unit identifier
# - t: time period
# - y_base, y_jump_m, y_smooth_m: outcome variables
# - z: policy variable
# - x_r: control/proxy variable

# Example usage with sample data
estimates <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_jump_m",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  post = 3,
  pre = 0,
  kernel = "fixest"
)

EventStudyPlot(estimates = estimates)
```

--------------------------------

### example_data - Sample Panel Dataset

Source: https://context7.com/jmslab/eventstudyr/llms.txt

A sample panel dataset from Freyaldenhoven et al. (2021) used for demonstrating event study analysis.

```APIDOC
## example_data

### Description
A sample panel dataset containing 2000 rows and 12 columns, including unit identifiers, time periods, outcome variables, policy indicators, and control variables.

### Key Variables
- **id** (integer) - Unit identifier.
- **t** (integer) - Time period.
- **y_base, y_jump_m, y_smooth_m** (numeric) - Outcome variables.
- **z** (integer) - Policy variable.
- **x_r** (numeric) - Control/proxy variable.
```

--------------------------------

### Estimate Dynamic OLS Model with Controls

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Fit a dynamic OLS model with anticipation effects, controls, unit and time fixed effects, and clustered standard errors. The 'fixest' kernel is utilized, and anticipation effects normalization is enabled.

```r
library(eventstudyr)

# Dynamic OLS model with anticipation effects and controls
estimates_dynamic <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_base",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  controls = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 3,
  overidpost = 5,
  pre = 2,
  overidpre = 4,
  normalize = -3,
  cluster = TRUE,
  kernel = "fixest",
  anticipation_effects_normalization = TRUE
)

summary(estimates_dynamic$output)
```

--------------------------------

### Estimate Static Model

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Fit a static event study model, which does not consider event-time dynamics. This is useful for a simple estimation of the overall policy effect.

```r
library(eventstudyr)

# Static model (no event-time dynamics)
estimates_static <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_jump_m",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  FE = TRUE,
  TFE = TRUE,
  post = 0,
  overidpost = 0,
  pre = 0,
  overidpre = 0,
  cluster = TRUE,
  kernel = "fixest"
)

summary(estimates_static$output)
```

--------------------------------

### Create Basic Event Study Plot

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Generate a publication-quality event study plot using EventStudyPlot. This function takes the output from EventStudy and creates a ggplot2 object with default confidence intervals and sup-t bands.

```r
library(eventstudyr)
library(ggplot2)
set.seed(1234)  # For reproducible sup-t bands

# Estimate model
estimates <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_smooth_m",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  controls = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 3,
  overidpost = 5,
  pre = 2,
  overidpre = 4,
  normalize = -3,
  kernel = "fixest"
)

# Basic event study plot with default settings
plt <- EventStudyPlot(estimates = estimates)
plt
```

--------------------------------

### Plot Event Study with Smoothest Confound Path

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Visualize the event study results along with the smoothest confound path by setting the 'smpath' argument to TRUE in EventStudyPlot. This helps in assessing potential confounding factors.

```r
library(eventstudyr)
library(ggplot2)
set.seed(1234)  # For reproducible sup-t bands

# Estimate model
estimates <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_smooth_m",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  controls = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 3,
  overidpost = 5,
  pre = 2,
  overidpre = 4,
  normalize = -3,
  kernel = "fixest"
)

# Plot with smoothest confound path
EventStudyPlot(estimates = estimates, smpath = TRUE)
```

--------------------------------

### TestLinear() - Hypothesis Testing

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Performs linear hypothesis tests on event study coefficients, including pre-trends and leveling-off effects using F-tests.

```APIDOC
## TestLinear(estimates, pretrends, leveling_off, test, test_name)

### Description
Performs linear hypothesis tests on event study coefficients. It evaluates whether coefficients meet specified hypotheses and returns a data frame with test statistics and p-values.

### Parameters
- **estimates** (object) - Required - The output object from the EventStudy() function.
- **pretrends** (boolean) - Optional - If TRUE, performs a test for pre-trends.
- **leveling_off** (boolean) - Optional - If TRUE, performs a test for leveling-off effects.
- **test** (string) - Optional - A string specifying a custom linear hypothesis (e.g., "z_fd_lag1 = z_fd").
- **test_name** (string) - Optional - A label for the custom hypothesis test.

### Response
- **data.frame** - Returns a data frame containing the Test name, F-statistic, and p-value.
```

--------------------------------

### Estimate Basic OLS Event Study

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Perform a basic OLS event study estimation using the EventStudy function. Specify outcome variable, policy variable, identifiers, time variable, and event window. The 'fixest' kernel is used for estimation.

```r
library(eventstudyr)

# Basic OLS event study estimation
estimates_ols <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_base",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  pre = 0,
  post = 3,
  normalize = -1,
  kernel = "fixest"
)

# Access estimated model and coefficients
summary(estimates_ols$output)
fixest::coeftable(estimates_ols$output)

# Access arguments used in estimation
estimates_ols$arguments$eventstudy_coefficients
```

--------------------------------

### Perform Linear Hypothesis Tests with TestLinear

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Evaluate event study coefficients using F-tests for pre-trends, leveling-off effects, or custom linear hypotheses.

```r
library(eventstudyr)

# Estimate model
estimates <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_base",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  controls = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 3,
  pre = 2,
  overidpre = 4,
  overidpost = 5,
  normalize = -3,
  cluster = TRUE,
  anticipation_effects_normalization = TRUE,
  kernel = "fixest"
)

# Test for pre-trends and leveling-off (default behavior)
test_results <- TestLinear(
  estimates,
  pretrends = TRUE,
  leveling_off = TRUE
)
print(test_results)
# Output: data.frame with Test, F-statistic, and p-value columns

# Custom hypothesis test on specific coefficients
TestLinear(
  estimates,
  test = "z_fd_lag1 = z_fd",
  test_name = "Custom Hypothesis Test",
  pretrends = TRUE,
  leveling_off = TRUE
)

# Test only for pre-trends
TestLinear(
  estimates,
  pretrends = TRUE,
  leveling_off = FALSE
)

# Test only for leveling-off
TestLinear(
  estimates,
  pretrends = FALSE,
  leveling_off = TRUE
)
```

--------------------------------

### Customize Event Study Plots with ggplot2

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Apply standard ggplot2 functions to modify the appearance of event study plots generated by the package.

```r
plt +
  geom_point(color = "red") +
  geom_hline(color = "gray", yintercept = 0) +
  theme_light() +
  theme(panel.grid.minor.x = element_blank())
```

--------------------------------

### Plot Event Study Estimates

Source: https://github.com/jmslab/eventstudyr/blob/main/README.md

Generates a plot of the estimated event study coefficients. This function takes the results from the EventStudy function as input.

```r
plt <- EventStudyPlot(estimates = estimates_ols)
plt
```

--------------------------------

### Estimate OLS Event Study Model

Source: https://github.com/jmslab/eventstudyr/blob/main/README.md

Estimates a linear panel event study model using OLS. Requires specifying outcome, policy, id, and time variables, along with control variables and the pre/post periods. Uses the 'fixest' kernel for estimation.

```r
library(eventstudyr)
set.seed(10) # for reproducibility of sup-t bands

estimates_ols <- EventStudy(
   estimator = "OLS",
   data = example_data,   # Use package sample data
   outcomevar = "y_smooth_m",
   policyvar = "z",
   idvar = "id",
   timevar = "t",
   controls = "x_r",
   pre = 0,  post = 4,
   kernel = "fixest"
)
```

--------------------------------

### Plot Event Study Without Sup-t Bands

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Generate an event study plot that excludes sup-t bands, displaying only confidence intervals. This can be achieved by setting the 'supt' argument to NULL in EventStudyPlot.

```r
library(eventstudyr)
library(ggplot2)
set.seed(1234)  # For reproducible sup-t bands

# Estimate model
estimates <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_smooth_m",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  controls = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 3,
  overidpost = 5,
  pre = 2,
  overidpre = 4,
  normalize = -3,
  kernel = "fixest"
)

# Plot without sup-t bands (only confidence intervals)
EventStudyPlot(estimates = estimates, supt = NULL)
```

--------------------------------

### Plot Event Study with Mean and Zero Line

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Enhance the event study plot by adding the mean of the outcome variable at the normalization period and a zero effect line. This provides additional context for interpreting the estimated effects.

```r
library(eventstudyr)
library(ggplot2)
set.seed(1234)  # For reproducible sup-t bands

# Estimate model
estimates <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_smooth_m",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  controls = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 3,
  overidpost = 5,
  pre = 2,
  overidpre = 4,
  normalize = -3,
  kernel = "fixest"
)

# Plot showing mean of outcome variable at normalization period
EventStudyPlot(
  estimates = estimates,
  add_mean = TRUE,
  add_zero_line = TRUE
)
```

--------------------------------

### Customize Event Study Plot Appearance

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Customize the appearance of an event study plot by setting axis titles and y-axis breaks. This allows for tailored visualization of the estimated effects.

```r
library(eventstudyr)
library(ggplot2)
set.seed(1234)  # For reproducible sup-t bands

# Estimate model
estimates <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_smooth_m",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  controls = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 3,
  overidpost = 5,
  pre = 2,
  overidpre = 4,
  normalize = -3,
  kernel = "fixest"
)

# Customized plot with axis titles and y-axis breaks
EventStudyPlot(
  estimates = estimates,
  xtitle = "Relative time",
  ytitle = "Effect on outcome",
  ybreaks = seq(-2, 1, 0.5)
)
```

--------------------------------

### Estimate IV Event Study using FHS Method

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Perform an Instrumental Variables (IV) event study estimation using the Freyaldenhoven-Hansen-Shapiro (FHS) method. This includes unit and time fixed effects, clustered standard errors, and specified event windows.

```r
library(eventstudyr)

# Instrumental Variables estimation using FHS method
estimates_iv <- EventStudy(
  estimator = "FHS",
  data = example_data,
  outcomevar = "y_base",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  proxy = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 2,
  overidpost = 1,
  pre = 0,
  overidpre = 3,
  normalize = -1,
  cluster = TRUE,
  kernel = "fixest"
)

summary(estimates_iv$output)
```

--------------------------------

### Plot Event Study Disabling Hypothesis Tests

Source: https://context7.com/jmslab/eventstudyr/llms.txt

Disable the display of hypothesis testing results in the plot caption by setting 'pre_event_coeffs' and 'post_event_coeffs' to FALSE in EventStudyPlot. This simplifies the plot's annotation.

```r
library(eventstudyr)
library(ggplot2)
set.seed(1234)  # For reproducible sup-t bands

# Estimate model
estimates <- EventStudy(
  estimator = "OLS",
  data = example_data,
  outcomevar = "y_smooth_m",
  policyvar = "z",
  idvar = "id",
  timevar = "t",
  controls = "x_r",
  FE = TRUE,
  TFE = TRUE,
  post = 3,
  overidpost = 5,
  pre = 2,
  overidpre = 4,
  normalize = -3,
  kernel = "fixest"
)

# Disable hypothesis testing in plot caption
EventStudyPlot(
  estimates = estimates,
  pre_event_coeffs = FALSE,
  post_event_coeffs = FALSE
)
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.