AUTOBOX.RPF

AUTOBOX.RPF uses @BJDIFF and @GMAUTOFIT to do an automated choice for the differencing and seasonal ARIMA specification for a series, then uses BOXJENK to estimate the chosen model, allowing for automated outlier detection.

Full Program

cal(m) 1992:1
open data x12test.xls
data(format=xls,org=columns) 1992:1 2008:7 u11bvs
set ldata = log(u11bvs)
*
@bjdiff(diffs=2,sdiffs=1) ldata
*
@gmautofit(diffs=%%autod,sdiffs=%%autods,const=%%autoconst,report) ldata
*
boxjenk(diffs=%%autod,sdiffs=%%autods,const=%%autoconst,$
ar=%%autop,sar=%%autops,ma=%%autoq,sma=%%autoqs,$
outliers=standard) ldata

Output

BJDiff Table, Series LDATA

Reg Diff Seas Diff Intercept Crit

0 0 No 0.832203

0 0 Yes -5.059814

0 1 No -6.304497

0 1 Yes -6.377040

1 0 No -5.985028

1 0 Yes -5.969354

1 1 No -6.524186*

1 1 Yes -6.496244

2 0 No -5.487755

2 0 Yes -5.461052

2 1 No -6.121989

2 1 Yes -6.093772

Search for Minimum BIC Model

Series LDATA

with 1 regular and 1 seasonal differences

AR MA AR(s) MA(s) LogL BIC

3 0 0 0 344.2921867 -672.907133

3 0 0 1 366.9063632 -712.909740

3 0 1 0 359.3487679 -697.794549

3 0 1 1 366.9723395 -707.815946

0 0 0 1 363.0333916 -720.841036

0 1 0 1 366.5535644 -722.655635

0 2 0 1 366.6361541 -717.595068

1 0 0 1 366.6299545 -722.808416

1 1 0 1 366.6053724 -717.533505

1 2 0 1 366.6664315 -712.429876

2 0 0 1 366.6347398 -717.592240

2 1 0 1 366.6319549 -712.360923

2 2 0 1 378.4816368 -730.834540

2 2 0 0 351.4513044 -681.999622

2 2 0 1 378.4816368 -730.834540*

2 2 1 0 368.3978326 -710.666932

2 2 1 1 378.6693394 -725.984199

Forward Addition pass 1

Largest t-statistic is LS(1999:04)= -3.506 < 3.942 in abs value

Box-Jenkins - Estimation by ML Gauss-Newton

Convergence in 34 Iterations. Final criterion was 0.0000096 <= 0.0000100

Dependent Variable LDATA

Monthly Data From 1993:02 To 2008:07

Usable Observations 186

Degrees of Freedom 181

Centered R^2 0.9557586

R-Bar^2 0.9547809

Uncentered R^2 0.9999861

Mean of Dependent Variable 8.6000977760

Std Error of Dependent Variable 0.1530033261

Standard Error of Estimate 0.0325358433

Sum of Squared Residuals 0.1916031787

Log Likelihood 372.2421

Durbin-Watson Statistic 1.9841

Q(36-5) 58.5000

Significance Level of Q 0.0020193

Variable Coeff Std Error T-Stat Signif

************************************************************************************

1. AR{1} 0.518827183 0.721027058 0.71957 0.47271979

2. AR{2} 0.240830385 0.539723705 0.44621 0.65597854

3. MA{1} -0.799230271 0.737886595 -1.08313 0.28018960

4. MA{2} -0.175975660 0.709647874 -0.24798 0.80443405

5. SMA{12} -0.574164273 0.072251715 -7.94672 0.00000000