Leopoldo Catania and Anna Gloria Billé, "Dynamic Spatial Autoregressive Models with Autoregressive and Heteroskedastic Disturbances", Journal of Applied Econometrics, Vol. 32, No. 6, 2017, pp. 1178-1196. This readme file documents the data set used in the paper. The data set is confidential but available from Datastream. There are three data sets used in the paper: 1) the series of sectoral returns, which is a panel of 18 series 2) the series of the exogeneous regressor 3) the series used for the construction of the weighting matrices For each data set, we report the RICs code (RIC), the series name (NAME) as displayed by Datastream, and the label (LABEL) used in the paper. Each RIC code identifies a series. RIC NAME LABEL .DJUSEN DJ US Oil&Gas EN .DJUSCH DJ CHEMICALS CH .DJUSBS DJ US BASIC RCS BS .DJUSCN DJ US ConsMatrls CN .DJUSIG DJ US INDS GOOD IG .DJUSAP DJ US AutoParts AP .DJUSFB DJ FOOD BEV FB .DJUSNG DJ US PerHldGds NG .DJUSHC DJ US HlthCare HC .DJUSME DJ US MEDIA NDX ME .DJUSCG DJ US TravLeis CG .DJUSTL DJ TELECOMM TL .DJUSUT DJ UTILITIES UT .DJUSBK DJ BANKS BK .DJUSIR DJ INSURANCE IR .DJUSRE DJ REAL EST RE .DJUSFI DJ US FINL SVCS FI .DJUSTC DJ TECHNOLOGY TC The RIC for the series of the exogeneous regressor (S&P500 log returns) is .SPX The weighting matrices are constructed using financial indicators associated to the sectoral indices. The instruments used to download the series are : Instrument LABEL TR.Index_MKT_CAP_USD_RTRS MKT TR.Index_PRICE_TO_BOOK_RTRS PB TR.Index_PE_RTRS PE TR.Index_DIV_YLD_RTRS DY