In EventStudyTools/api-wrapper.r: Event Study Analysis

knitr::opts_chunk$set(echo = TRUE)

Motivation {#motivation}

You can manually adjust all parameters in your R environment. Save this object and use it in all your Event Studies. This vignette gives you an overview on how to set the correct parameters in R.

Options {#options}

You are able to set some general options in the R option object. This enables you to setup your API credentials for all R sessions.

• EventStudy.URL: The API url (default: see bottom)
• EventStudy.KEY: The API key (default: NULL)
• EventStudy.verbose: Verbose level (range: 1 - 3, default: 3)
• EventStudy.tryAttempts: Attempts to retry http request (default: 3)

The options can be changed by the options function in R, e.g.

options(EventStudy.URL = "http://api.eventstudytools.com")

The options can be retained by the

getOption("EventStudy.tryAttempts")

Performing an Event Study (ARC, AVC, or AVyC) {#performing-an-event-study-arc-avc-or-avyc}

After you have set your parameters, you can easily perform your Event Study:

apiUrl <- "http://api.eventstudytools.com"
apiKey <- "Please insert your key here"

# Setup API Connection
estSetup <- EventStudyAPI$new(apiUrl)
estSetup$authentication(apiKey)

# Perform Event Study
estResults <- estSetup$performEventStudy(estParams     = volatilityEstParams, 
                                         dataFiles     = c("request_file" = "01_RequestFile.csv",
                                                           "firm_data"    = "02_firmData.csv",
                                                           "market_data"  = "03_marketData.csv"),
                                         downloadFiles = T, # download result files
                                         checkFiles    = T) # check input files                     

When you set your url or api key by EventStudy.URL and EventStudy.KEY then just leave input parameters in .$new() and .$authentication() empty. A default key and url can be set by:

options(EventStudy.KEY = apiKey)

General Event Study Parameters {#general-event-study-parameters}

There are several parameters applicable to all types of Event Studies, including Return, Volatility, and Volume Event Studies. Each Event Study type requires initializing its own R6 object before performing the analysis:

• ARCApplicationInput: Return Event Study
• AVCApplicationInput: Volume Event Study
• AVyCApplicationInput: Volatility Event Study

For the remainder of this section, we will focus on a Return Event Study object. However, the parameters discussed in this section can also be applied to the other two objects.

returnEstParams <- ARCApplicationInput$new()

Return File Type {#return-file-type}

The return type defines the type of the result file. Possible options are

• csv (default): comma-separated file with ; as delimiter,
• xls: Excel 2007,
• xlsx: Excel 2010 and newer, or
• ods: OpenOffice Spreadsheet.

The R API just works with CSV files.

Usage:

returnEstParams$setResultFileType("xlsx")

Return Type {#return-type}

• log (default): Return calculation by using log
• simple: Simple return calculation

Usage:

returnEstParams$setReturnType("simple")

Adjustment of Non-trading Days {#adjustment-of-non-trading-days}

• earlier: Select the previous trading day (Choose an earlier date if the event day falls on a non-trading day, such as a weekend).
• later (default): Retain non-trading days (Preserve announcement dates even if they occur on non-trading days, such as weekends).
• keep: Retain non-trading days (Preserve announcement dates even if they occur on non-trading days, such as weekends).
• skip: Omit specific observations (Exclude observations with event dates that fall on non-trading days, such as weekends).

Usage:

returnEstParams$setNonTradingDays("earlier")

Abnormal Return Event Study {#abnormal-return-event-study}

For performing Return Event Studies you need to initialize an ARCApplicationInput R6 object.

returnEstParams <- ARCApplicationInput$new()

Benchmark Models {#benchmark-models}

A comparison of our models can be found on our website.

• mm (default): Market Model\ The Market Model is a statistical method used to estimate the relationship between a stock's returns and overall market returns, often represented by a market index, through a linear regression model.
• mm-sw: Scholes/Williams Model\ The Scholes/Williams Model is a modified version of the Market Model that adjusts for potential biases in beta estimation caused by non-synchronous trading or infrequent trading of stocks by incorporating both leading and lagging market returns.
• cpmam: Comparison Period Mean Adjusted\ Comparison Period Mean Adjusted is a method used in event studies to calculate abnormal returns by adjusting each stock's return during the event window using the average return of the stock during a specified comparison period.
• ff3fm: Fama-French 3 Factor Model\ The Fama-French 3 Factor Model is an asset pricing model that expands upon the Capital Asset Pricing Model (CAPM) by incorporating two additional factors, size (market capitalization) and value (book-to-market ratio), to better explain stock returns and portfolio performance.
• ffm4fm: Fama-French-Momentum 4 Factor Model\ The Fama-French-Momentum 4 Factor Model is an extension of the Fama-French 3 Factor Model that adds a momentum factor, capturing the tendency of stocks with high past returns to continue outperforming, thus providing a more comprehensive framework for explaining stock returns and portfolio performance.
• garch: GARCH (1, 1) Model\ The GARCH (1, 1) Model, or Generalized Autoregressive Conditional Heteroskedasticity model, is a time-series forecasting model that accounts for volatility clustering by estimating the variance of a financial variable based on its own past values and past errors, thereby improving the accuracy of forecasts in financial markets.
• egarch: EGARCH (1, 1) Model\ The EGARCH (1, 1) Model, or Exponential Generalized Autoregressive Conditional Heteroskedasticity model, is an extension of the GARCH model that accounts for both positive and negative shocks to volatility by using a natural logarithm transformation, allowing for better modeling of asymmetric volatility effects in financial markets.

Usage:

returnEstParams$setBenchmarkModel("mm-sw")

Test Statistics {#test-statistics}

Per default all test statistics are activated. A detailed description can be found on our website.

Parametric Test Statistics:

• t-Test:
  • art
  • cart
  • abhart
• Cross-Sectional Test (CSect T):
  • aart
  • caart
  • abhart
• Time-Series Standard Deviation or Crude Dependence Test (CDA T):
  • aarcdat
  • caarcdat
• Patell or Standardized Residual Test (Patell Z):
  • aarptlz
  • caarptlz
• Kolari and Pynnönen adjusted Patell or Standardized Residual Test (Adjusted Patell Z):
  • aaraptlz
  • caaraptlz
• Standardized Cross-Sectional or BMP Test (StdCSect Z):
  • aarbmpz
  • caarbmpz
• Kolari and Pynnönen Adjusted Standardized Cross-Sectional or BMP Test (Adjusted StdCSect Z)
  • aarabmpz
  • caarabmpz
• Skewness Corrected Test (Skewness Corrected T):
  • aarskewadjt
  • caarskewadjt
  • abharskewadjt
• Jackknife Test (Abbr.: Jackknife T)
  • aarjackknivet
  • caarjackknivet

Non-Parametric Test Statistics:

• Corrado Rank Test (Rank Z):
  • aarrankz
  • caarrankz
• Generalized Rank T Test (Generalized Rank T):
  • aargrankt
  • caargrankt
• Generalized Rank Z Test (Generalized Rank Z):
  • aargrankz
  • caargrankz
• Cowan Generalized Sign Test (Generalized Sign Z)
  • aargsignz
  • caargsignz

Usage:

returnEstParams$setTestStatistics(c("aarptlz", "aarrankz"))

Abnormal Volume Event Study {#abnormal-volume-event-study}

For performing Return Event Studies you need to initialize an AVCApplicationInput R6 object.

volumeEstParams <- AVCApplicationInput$new()

Benchmark Models {#benchmark-models-1}

• mm (default): Market Model
• mm-sw: Scholes/Williams Model
• cpmam: Comparison Period Mean Adjusted
• ff3fm: Fama-French 3 Factor Model
• ffm4fm: Fama-French-Momentum 4 Factor Model

Usage:

volumeEstParams$setBenchmarkModel("mm-sw")

Test Statistics {#test-statistics-1}

Per default all test statistics are activated. A detailed description can be found on our website.

Parametric Test Statistics:

• t-Test:
  • art
  • cart
  • abhart
• Cross-Sectional Test (CSect T):
  • aart
  • caart
  • abhart
• Time-Series Standard Deviation or Crude Dependence Test (CDA T):
  • aarcdat
  • caarcdat
• Patell or Standardized Residual Test (Patell Z):
  • aarptlz
  • caarptlz
• Kolari and Pynnönen adjusted Patell or Standardized Residual Test (Adjusted Patell Z):
  • aaraptlz
  • caaraptlz
• Standardized Cross-Sectional or BMP Test (StdCSect Z):
  • aarbmpz
  • caarbmpz
• Kolari and Pynnönen Adjusted Standardized Cross-Sectional or BMP Test (Adjusted StdCSect Z)
  • aarabmpz
  • caarabmpz
• Skewness Corrected Test (Skewness Corrected T):
  • aarskewadjt
  • caarskewadjt
  • abharskewadjt
• Jackknife Test (Abbr.: Jackknife T)
  • aarjackknivet
  • caarjackknivet

Non-Parametric Test Statistics:

• Corrado Rank Test (Rank Z):
  • aarrankz
  • caarrankz
• Generalized Rank T Test (Generalized Rank T):
  • aargrankt
  • caargrankt
• Generalized Rank Z Test (Generalized Rank Z):
  • aargrankz
  • caargrankz
• Cowan Generalized Sign Test (Generalized Sign Z)
  • aargsignz
  • caargsignz

Usage:

volumeEstParams$setTestStatistics(c("aarptlz", "aarrankz"))

Abnormal Volatility Event Study {#abnormal-volatility-event-study}

For performing Return Event Studies you need to initialize an AVyCApplicationInput R6 object.

volatilityEstParams <- AVyCApplicationInput$new()

Benchmark Models {#benchmark-models-2}

• garch (default): Garch(1, 1) Model

Test Statistics {#test-statistics-2}

Per default all test statistics are activated. A detailed description can be found on our website.

Volatility Test Statistics:

• Cross-Sectional Test for Abnormal Volatility
  • aarcsvyt
• Corrected Cross-Sectional Test for Abnormal Volatility
  • aarcscvyt
• AAR Volatility Ratio-Test
  • aarlambdaz

Abnormal Return Test Statistics:

• Cross-Sectional Test for Abnormal Return
  • aarcsart
• Cross-Sectional-Corrected Test for Abnormal Return
  • aarcscart

Usage:

volatilityEstParams$setTestStatistics(c("aarptlz", "aarrankz"))

How to Cite {#how-to-cite}

Please cite our work in your publication.

EventStudyTools/api-wrapper.r documentation built on April 16, 2023, 1:04 p.m.