event2car_range: Cumulative abnormal returns for firm(s) at event(s)

In LisaLechner/event2car: Calculates cumulative abnormal return(s) of event(s) for firm(s)

Description

The function calculates cumulative abnormal returns for a specific time-range before and after the event(s).

Usage

event2car_range(
  returns = NULL,
  regressor = NULL,
  event_date = NULL,
  method = c("mean_adj", "mrkt_adj_within", "mrkt_adj_out"),
  imputation_returns = c("approx", "mean", "drop", "pmm"),
  imputation_regressor = c("approx", "mean"),
  car_lag = 1,
  car_lead = 5,
  estimation_period = 150,
  range_before = 30,
  range_after = 30,
  output_format = c("zoo", "data.frame")
)

Arguments

returns	an object of class zoo containing rates of returns of securities.
regressor	an object of the same class as returns containing regressors. The argument can be omitted, if market model is mean_adj.
event_date	a character object or an object of class Date containing one event date or multiple event dates in the format YYYY-MM-DD.
method	a character indicating the method used to calculate abnormal returns and cumulative abnormal returns. Choose among mean_adj, mrkt_adj_out, and mrkt_adj_within.
imputation_returns	a character indicating the way of dealing with missing values in returns data: drop: No imputation at all and dropping observations with NAs, mean: imputing missing data with the mean stock return (for further information see '?zoo::na.aggregate') , approx: imputing missing data using interpolated values (for further information see '?zoo::na.approx'), pmm: imputing missing data by dint of predictive mean matching used in the mice package (for further information see '?mice::mice').
imputation_regressor	a character indicating the way of dealing with missing values in regressor data: mean: imputing missing data with the mean stock return (for further information see '?zoo::na.aggregate') , approx: imputing missing data using interpolated values (for further information see '?zoo::na.approx'),
car_lag	an object of class intenger measuring the start of the event window. The default is 1 day prior to the event date.
car_lead	an object of class intenger measuring the end of the event window. The default is 5 days after the event date.
estimation_period	an object of class intenger stating the number of days prior to the event over which the market model parameters are estimated. Default is 150 days. Note that the event period itself is not included in the event period to prevent the event from influencing the normal performance model parameter estimates.
range_before	a numeric value indicating the time range to be included before the first event.
range_after	a numeric value indicating the time range to be included after the last event.
output_format	a character value signaling whether the cumulative abnormal returns time series should be returned as zoo or data.frame object.

Details

The package covers three models for the calculation of the cumulative abnormal returns: - Mean-adjusted model (mean_adj) - Market-adjusted model using out-of-sample estimation (mrkt_adj_out) - Market-adjusted model using within-sample estimation (mrkt_adj_within) This is the logic suggested by multiple scholars. See references below.

The generic function is dispatched for such classes as zoo. (future versions of the package allow for classes of data.frame.)

If method is mrkt_adj_out or mrkt_adj_within and regressor has the length greater than one, the first element of regressor will be applied for each security in returns.

Value

an object of class event2car_range which contains the cumulative abnormal returns time series (car_timeseries), information on event_dates, information on firms, and information on method, imputation_returns, and imputation_regressor.

References

MacKinlay, A.C. Event Studies in Economics and Finance. Journal of Economic Literature, 35(1):13-39, 1997.

Brown S.J., Warner J.B. Using Daily Stock Returns, The Case of Event Studies. Journal of Financial Economics, 14:3-31, 1985.

Davies, R., Studnicka, Z. The heterogeneous impact of Brexit: Early indications from the FTSE. European Economic Review, 110:1-17, 2018.

Examples

trumpelection <- as.Date("2016-11-08")
returns_firms <- tech_returns[,2:19]
return_indx <- tech_returns[,1]
effect_trump <- event2car_range(returns=returns_firms,regressor=return_indx,
                                imputation_returns="mean",
                                event_date=trumpelection,method="mean_adj")

LisaLechner/event2car documentation built on Dec. 24, 2021, 8:48 a.m.