M1Germany: German M1 Money Demand
In dynlm: Dynamic Linear Regression

Description Usage Format Details Source References See Also Examples

German M1 money demand.

1	data(M1Germany)

M1Germany is a "zoo" series containing 4 quarterly time series from 1960(1) to 1996(3).

logm1: logarithm of real M1 per capita,
logprice: logarithm of a price index,
loggnp: logarithm of real per capita gross national product,
interest: long-run interest rate,

This is essentially the same data set as GermanM1, the important difference is that it is stored as a zoo series and not as a data frame. It does not contain differenced and lagged versions of the variables (as GermanM1) does, because these do not have to be computed explicitly before applying dynlm.

The (short) story behind the data is the following (for more detailed information see GermanM1): Lütkepohl et al. (1999) investigate the linearity and stability of German M1 money demand: they find a stable regression relation for the time before the monetary union on 1990-06-01 but a clear structural instability afterwards. Zeileis et al. (2005) re-analyze this data set in a monitoring situation.

The data is provided by the German central bank and is available online in the data archive of the Journal of Applied Econometrics http://qed.econ.queensu.ca/jae/1999-v14.5/lutkepohl-terasvirta-wolters/.

Lütkepohl H., Teräsvirta T., Wolters J. (1999), Investigating Stability and Linearity of a German M1 Money Demand Function, Journal of Applied Econometrics, 14, 511–525.

Zeileis A., Leisch F., Kleiber C., Hornik K. (2005), Monitoring Structural Change in Dynamic Econometric Models, Journal of Applied Econometrics, 20, 99–121.

GermanM1

data("M1Germany")
## fit the model of Luetkepohl et al. (1999) on the history period
## before the monetary unification 
histfm <- dynlm(d(logm1) ~ d(L(loggnp, 2)) + d(interest) + d(L(interest)) + d(logprice) +
                           L(logm1) + L(loggnp) + L(interest) +
                           season(logm1, ref = 4),
                data = M1Germany, start = c(1961, 1), end = c(1990, 2))

## fit on extended sample period
fm <- update(histfm, end = c(1995, 4))

if(require("strucchange")) {
  scus <- gefp(fm, fit = NULL)
  plot(scus, functional = supLM(0.1))
}