Matthieu Droumaguet, Anders Warne, Tomasz Woźniak, "Granger Causality and Regime Inference in Markov-Switching VAR Models with Bayesian Methods", Journal of Applied Econometrics, Vol. 32, No. 4, 2017, pp. 802-818. The data are contained in the file "data-dww.txt". It contains 646 monthly observations for the sample 1959M2-2012M11 of the two monthly U.S. time series of the M1 money stock and the industrial production index. Both variables are taken from the Citibase database. The data are seasonally adjusted, transformed using first differences of the natural logarithm, and multiplied by 1200. data-dww.txt is an ASCII file in DOS format. It is zipped in the file data-dww.zip. Unix/Linux users should use "unzip -a". The first row of the data-file contains the variable names, while the remaining rows contain date information and values of the variables. Name Description ----------------------- IP Seasonally adjusted industrial production index in logarithmic rates of return multiplied by 1200. M1 Seasonally adjusted M1 money stock in logarithmic rates of return multiplied by 1200. The underlying R code we have used to estimate the Markov-switching Vector Autoregressive models can be downloaded from: http://szeridan.blogspot.it/2016/05/bayesianMSVAR.html