John Maheu and Yong Song, "An Efficient Bayesian Approach to Multiple
Structural Change in Multivariate Time Series", Journal of Applied
Econometrics, Vol. 33, No. 2, 2018, pp. 251-270.

The data files are data1_VAR.csv and data2_oilGDP.csv. These are ASCII
files in DOS format, which are zipped in the file ms-data.zip.
Unix/Linux users should use "unzip -a".

For the first data set, data1_VAR.csv:

From 1959M2 to 2011M2 (625 observations)

col1: date
col2: unemployment rate
col3: Core PCE index(1200 * log difference)
col4: Nonfarm employment level (1200 * log difference),
col5: retail sales level(1200 * log difference)
col6: Housing starts level(100 * log difference)
col7: Industrial produce index(1200 * log difference)
col8: federal funds rate

Source: All data are from FRED. Their Series IDs are:

unemployment rate: M0892AUSM156SNBR
core PCE: PCEPILFE
Housing starts: HOUST
nonfarm employment: PAYEMS
retail sales: RSXFS
industrial production index: INDPRO
Federal Funds rate: FEDFUNDS


For the second data set, data2_oilGDP.csv:

1974Q2-2015Q2 (165 observations)

col1: date
col2: oil price
col3: real GDP growth

data sources:

oil price: composite refiner's acquisition cost (data label: EEPRPC)
from Citibase

real GDP growth rate: (chained, seasonally adjusted) is from Bureau of
Economic Analysis.