eventstudies (version 1.2.2)

eesInference: Extreme event study inference estimation

Description

This function performs event study analysis on extreme event dates (‘eesDates’) and using formatted output (‘get.clusters.formatted’)

Usage

eesInference(input, event.lists, event.window, to.remap = TRUE, remap = "cumsum", 
   		inference = "TRUE", inference.strategy = "bootstrap")

Arguments

input

a formatted cluster object, as returned by ‘get.clusters.formatted’ function.

event.lists

a ‘list’ of normal and purged events as returned by ‘eesDates’.

event.window

an ‘integer’ of length 1 that specifies a symmetric event window around the event date.

to.remap

‘logical’, indicating whether or not to remap the data in ‘input’.The default setting is ‘TRUE’

remap

‘character’, indicating the type of remap required, “cumsum”, “cumprod”, or “reindex”. Used when ‘to.remap’ is ‘TRUE’.

inference

‘logical’, specifying whether to undertake statistical inference and compute confidence intervals. The default setting is ‘TRUE’.

inference.strategy

a ‘character’ scalar specifying the inference strategy to be used for estimating the confidence interval. Presently, two methods are available: “bootstrap” and “wilcox”. The default setting is ‘bootstrap’.

Value

Format of event study output is a ‘matrix’ containing mean or median estimate with confidence interval; ‘NULL’ if there are no “success” “outcomes”. See phys2eventtime for more details.

A ‘list’ with class attribute “ees” holding the following four event study output elements:

good.normal

an event study inference ‘matrix’ for right tail unclustered events, termed as normal

bad.normal

an event study inference ‘matrix’ for left tail unclustered events, termed as normal

good.purged

an event study inference ‘matrix’ for right tail clustered and unclustered events, termed as purged

bad.purged

an event study inference ‘matrix’ for left tail clustered and unclustered events, termed as purged

Details

This function performs event study analysis using eventstudy function on the extreme event dates of normal (unclustered events) and purged (clustered and unclustered events) sets. These interesting dates are obtained from function ‘eesDates’. The function can estimate confidence interval using different inference strategies as provided by eventstudy().

The function does not do market model adjustment but takes the output of get.clusters.formatted as it's input.

References

Ila Patnaik, Nirvikar Singh and Ajay Shah (2013). Foreign Investors under stress: Evidence from India. International Finance, 16(2), 213-244. http://onlinelibrary.wiley.com/doi/10.1111/j.1468-2362.2013.12032.x/abstract http://macrofinance.nipfp.org.in/releases/PatnaikShahSingh2013_Foreign_Investors.html

Examples

# NOT RUN {
data(OtherReturns)

formattedClusters <- get.clusters.formatted(event.series = OtherReturns[, "SP500"], 
                                           response.series = OtherReturns[, "NiftyIndex"])

event.lists <- eesDates(formattedClusters)

inference <- eesInference(input = formattedClusters,
                          event.lists = event.lists,
                          event.window = 5)
str(inference, max.level = 2)
# }