                                                          9 February 1998

Example 8:  This example uses the "Unit Auto Sales" series of the 
            U.S. Bureau of Labor Statistics.  The spec file 
            beauto.spc is for the final "hybrid" model from
            Section 5.4 of the article "New Capabilities and
            Methods of the X-12-ARIMA Seasonal Adjustment Program"
            by D.F. Findley, B.C. Monsell, W.R. Bell, M.C. Otto
            and B.C. Chen that appeared in the Journal of Business and
            Economic Statistics (1998, 16:127-177).  Its AIC
            history output from the history spec shows the value
            of using the "subjective" preadjustments of the effects 
            of automobile manufacturers' rebate campaigns to promote 
            sales (in beauto.v3).  

            Compare the AIC history before 1985 from this run 
            with the AIC history from beauto2.spc, which does not
            perform the preadjustment.  Graphs of the AIC history
            are available in X-12-Graph.



# Example 8:  beauto.spc

# This is the spec file for the final "hybrid" model. 
# The AIC history output from the history spec shows the value of using the
# "subjective" preadjustments (in beauto.v3) of the effects of automobile 
# manufacturers' rebates campaigns to promote sales on the "Unit Auto Sales"
# series of the U.S. Bureau of Labor Statistics.

series{
 title='BEA unit auto sales : with prior adjustments'
 start=1972.jan
 data=(
  6050  6927  7657  7683  8802  8693  7624  6503  7351  9255  8844  7137
  7290  7674  9547  8544  9625  9007  8009  6792  7488  8505  7722  5683
  5454  5615  6453  6935  7566  6868  6811  6600  5861  6218  5002  4235
  4576  5309  5170  5089  5920  6084  6267  5253  5818  7646  6465  5912
  5801  6409  8041  7768  7819  8169  7259  6068  6370  7227  7132  6857
  5928  6547  8804  8098  8213  9066  7200  7152  6465  8591  7274  6374
  5381  6196  8692  8488  9471  9313  7459  7401  6537  8741  7592  6368
  6356  6663  8526  7528  7869  6908  6796  6975  5928  7213  5996  5543
  5880  5915  6704  5410  4987  5111  5424  4868  4861  6642  5296  4716
  4700  5438  7192  5338  5243  5179  4974  6022  5189  4916  4321  3576
  3682  4572  5760  4989  5845  4519  4300  4094  4885  4845  5546  4439
  4141  4425  6003  5777  6300  6681  5769  5310  5383  6642  5905  5594
  5834  6550  7562  7211  8033  7274  6841  6035  5667  6896  6007  5607
  6280  6453  7689  7883  8076  6767  6335  7448  8394  5984  5158  5580
  6359  6134  6494  7196  7863  7363  6487  6733  9249  6328  5211  6733
  4278  5581  6834  6938  6223  6567  6110  6540  6132  5239  4860  5507
  5312  6491  7337  6516  7020  7224  6053  6030  5788  5917  5538  6168
  5121  5536  6417                                                      
 )
}
transform{
 function=log
 start=1972.jan
 file="beauto.adj"
 mode=ratio
 type=permanent
}
regression{
 variables=(const td AO1975.Feb AO1988.Dec)
 user=("User.Aug" "User.Sep" "User.Oct" "User.Nov" "User.D&J")
 start=1968.jan
 file="beauto.v3"
 usertype=ao
}
arima{
 model=(0 1 1)(0 1 1)
}
check{  }
x11{   }
history{
  start=1979.jan
  estimates=aic
  fixmdl=no
}



# Example 8:  beauto0.spc

# This is spec file for the model without preadjustments for the 
# "Unit Auto Sales" series of the U.S. Bureau of Labor Statistics.

series{
 title='BEA unit auto sales : no prior adjustments'
 start=1972.jan
 data=(
  6050  6927  7657  7683  8802  8693  7624  6503  7351  9255  8844  7137
  7290  7674  9547  8544  9625  9007  8009  6792  7488  8505  7722  5683
  5454  5615  6453  6935  7566  6868  6811  6600  5861  6218  5002  4235
  4576  5309  5170  5089  5920  6084  6267  5253  5818  7646  6465  5912
  5801  6409  8041  7768  7819  8169  7259  6068  6370  7227  7132  6857
  5928  6547  8804  8098  8213  9066  7200  7152  6465  8591  7274  6374
  5381  6196  8692  8488  9471  9313  7459  7401  6537  8741  7592  6368
  6356  6663  8526  7528  7869  6908  6796  6975  5928  7213  5996  5543
  5880  5915  6704  5410  4987  5111  5424  4868  4861  6642  5296  4716
  4700  5438  7192  5338  5243  5179  4974  6022  5189  4916  4321  3576
  3682  4572  5760  4989  5845  4519  4300  4094  4885  4845  5546  4439
  4141  4425  6003  5777  6300  6681  5769  5310  5383  6642  5905  5594
  5834  6550  7562  7211  8033  7274  6841  6035  5667  6896  6007  5607
  6280  6453  7689  7883  8076  6767  6335  7448  8394  5984  5158  5580
  6359  6134  6494  7196  7863  7363  6487  6733  9249  6328  5211  6733
  4278  5581  6834  6938  6223  6567  6110  6540  6132  5239  4860  5507
  5312  6491  7337  6516  7020  7224  6053  6030  5788  5917  5538  6168
  5121  5536  6417                                                      
 )
}
transform{
 function=log
}
regression{
 variables=(const td AO1975.Feb AO1988.Dec)
 user=("User.Aug" "User.Sep" "User.Oct" "User.Nov" "User.D&J")
 start=1968.jan
 file="beauto.v3"
 usertype=ao
}
arima{
 model=(0 1 1)(0 1 1)
}
check{  }
x11{   }
history{
  start=1979.jan
  estimates=aic
  fixmdl=no
}
