Forums  > Trading  > What is the best way to measure response for a shock?  
     
Page 1 of 1
Display using:  

hyqus


Total Posts: 1
Joined: Dec 2016
 
Posted: 2016-12-01 02:53
I am trying to measure the response to a shock in a financial time series but I am not sure what is the best way to actually measure the shock. For example, if I want to measure a shock compared to the recent history of the underlying time series, how much data do I use? How far back should I go? I am not sure how to determine the correct window size. Should it be a function of the response time being measured? Very appreciate any comments.

Tradenator


Total Posts: 1585
Joined: Sep 2006
 
Posted: 2016-12-04 13:31
Try improving clarity here. Shock to what? Measured in what terms: price, return, vol, debt? Once you put some precision around what your "shock" is then you can measure the response using the same metric, no? Presumably there is a sign flip between shock and response, I would think, but that may depend on your measure.

contango_and_cash


Total Posts: 72
Joined: Sep 2015
 
Posted: 2017-01-24 17:53
I have used this in SAS in the past.

The following exists in scipy.
Previous Thread :: Next Thread 
Page 1 of 1