Interface to lm for fitting exchange rate regression
models (Frankel-Wei models).
Usage
fxlm(formula, data, ...)
Arguments
formula
a "formula" describing the linear model to be fit.
For details see below.
data
a "zoo" time series.
...
arguments passed to lm.
Details
fxlm is a function for fitting exchange rate regression models also
known as Frankel-Wei models. It is a simple convenience interface to lm:
data is assumed to be a "zoo" series in which, by default, the
first column is the dependent variable. If formula is omitted, the first
column is regressed on the remaining columns in data. The main difference
compared to plain lm models is that the error variance is reported as
a full parameter (estimated by maximum likelihood) in the coef method
and the estfun method (but currently not in the vcov method).
Furthermore, the index (also known as the time stamps) of the underlying data set
can be extracted by the time/index method.
Value
An object of class "fxlm" inheriting from "lm".
References
Shah A., Zeileis A., Patnaik I. (2005), What is the New Chinese
Currency Regime?, Report 23, Department of Statistics and Mathematics,
Wirtschaftsuniversitaet Wien, Research Report Series, November 2005.
http://epub.wu.ac.at.
Zeileis A., Shah A., Patnaik I. (2010), Testing, Monitoring, and Dating Structural
Changes in Exchange Rate Regimes, Computational Statistics and Data Analysis,
54(6), 1696–1706. http://dx.doi.org/10.1016/j.csda.2009.12.005.
See Also
lm, fxregimes
Examples
## load package and data
library("fxregime")
data("FXRatesCHF", package = "fxregime")
## compute returns for CNY (and explanatory currencies)
## for one year after abolishing fixed USD regime
cny <- fxreturns("CNY", frequency = "daily",
start = as.Date("2005-07-25"), end = as.Date("2006-07-24"),
other = c("USD", "JPY", "EUR", "GBP"))
## estimate full-sample exchange rate regression
fm <- fxlm(CNY ~ USD + JPY + EUR + GBP, data = cny)
coef(fm)
summary(fm)
## test parameter stability (with double max test)
scus <- gefp(fm, fit = NULL)
plot(scus, aggregate = FALSE)
## which shows a clear increase in the variance in March 2006
## alternative tests: Andrews' supLM ...
plot(scus, functional = supLM(0.1))
## ... or Nyblom-Hansen test (Cramer-von Mises type test)
plot(scus, functional = meanL2BB)