9 February 1988 Example 4: Three series that are unsuitable for seasonal adjustment (m1gold.spc, m2airc.spc, and x1gold.spc), as various diagnostics in the output show. Look at the M and Q statistics and the sliding span diagnostics. Suggested graphs: overlay graph of the original and seasonally adjusted series, seasonal factors by month, and year on year graphs. Two of the spec files are listed below. # Example 4: m1gold.spc # This series, US Imports of Gold, has no stable seasonality. X-12-ARIMA # rejects adjustment of such a series. series {period=12 title=' ' file='m1gold.ori' name='m1gold' start=1979.01 print=(brief) span=(1980.01,) } transform {function=log} arima {model=(0 1 1)} check { } estimate {print=(brief)} outlier {types=all} x11 {mode=mult seasonalma=(s3x5) print=(brief +d8 +d9 +d13) } slidingspans{outlier=keep fixmdl=yes } forecast {maxlead=24 } # Example 4: x1gold.spc # This series, US Exports of Gold, has no stable seasonality. X-12-ARIMA # rejects adjustment of such a series. series {period=12 title='X-12-ARIMA run for x1gold' file='x1gold.ori' name='x1gold' start=1979.01 print=(brief) span=(1982.01,) } transform {function=log} regression {variables=(td1coef ao1986.3 ao1986.5 ao1986.6 ao1986.7) } arima {model=(0 1 1)(0 1 1)} check {print=all} outlier {print=brief} x11 {mode=mult seasonalma=(s3x9) print=(brief +d8 +d9 +d13) } slidingspans{outlier=keep fixmdl=yes } forecast {maxlead=24 }