[Eventstudies-discussion] Function eesPlot in package "eventstudies"
Rahul Choudhry
rahul.choudhry at gmail.com
Tue Nov 26 17:10:12 CET 2013
Dear Vikram & Ajay,
Thanks a lot for getting back to me. I have printed out the paper and will
go through it.
Vikram, I appreciate the explanation you have provided and how it relates
to what I am trying to accomplish. I had one quick clarification question
though. Since my events happens once every month (EOM when report is
released), whereas the FX rates (close rate) changes daily, the number of
elements in response time series > event time series.
In order to make the number of elements in the event time series vector =
response time series, do I just carry through the value until it changes
at the end of month.
So my z object to use in eesPlot would be something like following
Date Unemployment Rate ( Event)
FX
01-Nov-2013 5.5
1.5
02-Nov-2013 5.5
1.47
03-Nov-2013 5.5
1.6
.. 5.5
1.54
..
..
29-Nov-2013 5.5
1.6
30-Nov-2013 5.2
1.4
01-Dec-2013 ..(continues as 5.2 till 31-Dec-2013)
Thanks for your help again !
Rahul
On Tue, Nov 26, 2013 at 10:11 AM, Ajay Shah <ajayshah at mayin.org> wrote:
> Use
> http://macrofinance.nipfp.org.in/releases/PatnaikShahSingh2013_Foreign_Investors.htmlas the main page for the paper.
>
>
> On 26 November 2013 19:38, Vikram Bahure <economics.vikram at gmail.com>wrote:
>
>> Dear Rahul,
>>
>> Thank you for your note of appreciation.
>>
>> The 'eesPlot' function, replicates the Patnaik, Shah and Singh (2013)
>> paper, Foreign investors under stress Evidence from India<http://onlinelibrary.wiley.com/doi/10.1111/j.1468-2362.2013.12032.x/abstract>.
>> The paper uses modified event study methodology, which deals with clustered
>> events and purges out mixed events before performing event study.
>>
>> For instance, your response series is FX rate (FX) and event series is
>> Unemployment rate (UR). Methodology:
>> 1. Defining event: With the help of all the data, we get a distribution
>> of UR which is used to define event series. Extreme events on both sides of
>> the tail are used in event series and tail is defined using 'prob.value'
>> argument. ('prob.value=5' is 5% of tail on both sides).
>> 2. Mixed events: There can be cases when right tail event is next to left
>> tail event in an event series UR. With difficulty to interpret these
>> results, we discard them.
>> 3. Clustered events: After extracting tail events and discarding mixed
>> events, there can be cases where we find consecutive left or right tail
>> events. We fuse all the consecutive events into a single event and their
>> respective response series.
>> 4. Event study: Once we have modified event series (UR) and response
>> series (FX), event study graph is generated.
>>
>> I hope this was helpful. For more detailed understanding, I will suggest
>> you to download the paper.<http://onlinelibrary.wiley.com/doi/10.1111/j.1468-2362.2013.12032.x/abstract>
>>
>> Regards
>> Vikram Bahure
>> Consultant
>> NIPFP
>> New Delhi.
>>
>>
>> On Tue, Nov 26, 2013 at 10:54 AM, Rahul Choudhry <
>> rahul.choudhry at gmail.com> wrote:
>>
>>> Hi,
>>>
>>> First of all thanks a lot for putting this awesome package. I have been
>>> using it for the past one week or so for event profiling and looking at the
>>> impact of macro-economic events on various currencies.
>>>
>>> I was going through the package manual at the link below and trying to
>>> understand other functions in the package.
>>> http://cran.r-project.org/web/packages/eventstudies/eventstudies.pdf
>>>
>>> In particular I am trying to get my head around eesPlot function better
>>> and if I can use it in my analysis.
>>> I am trying to analyse the FX rates of G10 currencies (wrt USD) and how
>>> they behave with the release of US economic reports (eg Monthly
>>> unemployment numbers). In other words, my response series would be daily FX
>>> rate returns for a particular currency. However, when it comes to event
>>> series, I am not really sure of what I should be using.
>>>
>>> Following is a snippet of the file showing daily returns of different
>>> Forex rates
>>>
>>> Date EUR JPY CHF GBP CAD AUD NZD NOK SEK 1/3/2000 0 0 0 0 0 0 0 0 0
>>> 1/4/2000 0.002 -0.016 0.002 -0.001 -0.005 -0.009 -0.011 0.004 0.01
>>> 1/5/2000 0.002 -0.011 0.002 0.003 0.002 0.004 -0.005 0 0.001 1/6/2000
>>> -0.002 -0.01 0.005 0.003 -0.007 -0.004 -0.005 -0.001 -0.002 1/7/2000
>>> -0.002 0.002 -0.011 -0.006 0.002 0.001 0.003 -0.003 -0.006 1/10/2000
>>> -0.003 0.001 -0.003 -0.001 0 0.001 0.002 -0.007 -0.004 1/11/2000 0.007
>>> -0.007 0.006 0.007 0 0.003 -0.001 0.007 0.006 1/12/2000 -0.001 0 -0.002
>>> 0 0.002 0 0.004 -0.001 -0.002 1/13/2000 -0.005 -0.002 -0.006 -0.001
>>> 0.002 0.015 0.011 0.004 0
>>>
>>> Following is a snippet of Unemployment data. ( it is released at the
>>> end of every month). If I make a daily time series, every month from Day 1
>>> till end of month, I will be carrying the numbers from the end of last
>>> month when new data comes in.
>>>
>>> *Date* *Actual Line* *Survey Median Line* 01/31/2000 4 4 02/29/2000
>>> 4.1 4 03/31/2000 4.1 4 04/30/2000 3.9 4 05/31/2000 4.1 3.9
>>> 06/30/2000 4 4 07/31/2000 4 4 08/31/2000 4.1 4 09/30/2000 3.9 4.1
>>> 10/31/2000 3.9 4
>>> --
>>> Any assistance provided would be very helpful.
>>>
>>>
>>> Thanks for your time.
>>>
>>> Rahul
>>>
>>>
>>>
>>>
>>> Regards,
>>> Rahul
>>> Cell: 312 480 5469
>>>
>>> _______________________________________________
>>> Eventstudies-discussion mailing list
>>> Eventstudies-discussion at lists.r-forge.r-project.org
>>>
>>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/eventstudies-discussion
>>>
>>>
>>
>> _______________________________________________
>> Eventstudies-discussion mailing list
>> Eventstudies-discussion at lists.r-forge.r-project.org
>>
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/eventstudies-discussion
>>
>>
>
>
> --
> Ajay Shah
> ajayshah at mayin.org
> http://www.mayin.org/ajayshah
> http://ajayshahblog.blogspot.com
>
--
Regards,
Rahul
Cell: 312 480 5469
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/eventstudies-discussion/attachments/20131126/7db561dd/attachment.html>
More information about the Eventstudies-discussion
mailing list