[R-SIG-Finance] Removing Effect of Macroeconomic Variables from Time Series
Ajay Shah
@j@y@h@h @end|ng |rom m@y|n@org
Sun May 24 08:28:35 CEST 2020
There is one reasonably well established concept, of "augmented market
models" that are used to purge stock returns of the effects of the stock
market, the exchange rate, etc.
The R package for event studies <https://github.com/nipfpmf/eventstudies>
has a function lmAMM() which does this carefully, e.g. as shown in Patnaik
& Shah 2010
<http://macrofinance.nipfp.org.in/PDF/does-the-currency-regime-shape-unhedged-currency-exposure.pdf>.
Please be sure to fully understand and utilise the capabilities in makeX()
which has to be run before you get to lmAMM(). Many experienced researchers
have difficluties in published papers on these things, and wider use of
these concepts and the package will help.
The paper and package are written around *stock returns ~ market index +
exchange rate*. But the ideas and code are general and can be applied in
many other ways. Please do tell us about your experiences.
--
Ajay Shah
ajayshah using mayin.org
http://www.mayin.org/ajayshah
[[alternative HTML version deleted]]
More information about the R-SIG-Finance
mailing list