REPLICATION OF TIME-SERIES ANALYSIS OF BLAISDELL DATA (NKNW pp. 504-516)
USING COCHRANE-ORCUTT, HILDRETH-LU, & FIRST DIFFERENCES PROCEDURES

MON 3/22/99 9:10:04 AM

SYSTAT VERSION 7.0.1
COPYRIGHT (C) 1997, SPSS INC.

Welcome to SYSTAT!

>USE 'C:\SYSTAT7\S209\BLAIS.SYD'
SYSTAT Rectangular file C:\SYSTAT7\S209\BLAIS.SYD,
created Fri Mar 19, 1999 at 16:06:42, contains variables:
 Y            X

>mglh

>model y=constant+x

>save blaisres/resid data

>estimate
 
Dep Var: Y   N: 20   Multiple R: 0.999396   Squared multiple R: 0.998792
 
Adjusted squared multiple R: 0.998725   Standard error of estimate: 0.086056
 
Effect         Coefficient    Std Error     Std Coef Tolerance     t   P(2 Tail)
 
CONSTANT         -1.454750     0.214146     0.0        .       -6.79326  0.00000
X                 0.176283     0.001445     0.999396  1.000000  1.22E02  0.00000
 
                             Analysis of Variance
 
Source             Sum-of-Squares   df  Mean-Square     F-ratio       P
 
Regression            110.256878     1   110.256878 1.48881E+04    0.000000
Residual                0.133302    18     0.007406
 
Durbin-Watson D Statistic     0.735
First Order Autocorrelation   0.626
 
Residuals have been saved.
----------------------------------------------------------------------------------------------------------------------------------
Estimated rho (0.626) differs slightly from the one in NKNW (0.631166; Table 12.3 p. 510);
Estimated D-W statistic (0.735) is the same.
Next switch to file of residuals to do an index plot, then back to original
file 

>USE 'C:\SYSTAT7\S209\BLAISRES.SYD'
SYSTAT Rectangular file C:\SYSTAT7\S209\BLAISRES.SYD,
created Mon Mar 22, 1999 at 09:11:12, contains variables:
 ESTIMATE     RESIDUAL     LEVERAGE     COOK         STUDENT      SEPRED
 Y            X

>plot residual/stick ylimit=0 line dash=11
>USE 'C:\SYSTAT7\S209\BLAIS.SYD'
SYSTAT Rectangular file C:\SYSTAT7\S209\BLAIS.SYD,
created Fri Mar 19, 1999 at 16:06:42, contains variables:
 Y            X


>rem Cochrane-Orcutt procedure
>basic
File in use is C:\SYSTAT7\S209\BLAIS.SYD.
 
Variables in the SYSTAT Rectangular file are:
 Y            X
 BASIC statements cleared.

>let y1 = lag(y)

>let x1 = lag(x)

>let yp = y - 0.626*y1

>let xp = x - 0.626*x1

>run
SYSTAT file created.
 
  20 cases and 6 variables processed.
 BASIC statements cleared.
>page wide
>list x x1 xp y y1 yp
  Case number            X           X1           XP            Y           Y1           YP
        1       127.300000      .            .          20.960000      .            .
        2       130.000000   127.300000    50.310200    21.400000    20.960000     8.279040
        3       132.700000   130.000000    51.320000    21.960000    21.400000     8.563600
        4       129.400000   132.700000    46.329800    21.520000    21.960000     7.773040
        5       135.000000   129.400000    53.995600    22.390000    21.520000     8.918480
        6       137.100000   135.000000    52.590000    22.760000    22.390000     8.743860
        7       141.200000   137.100000    55.375400    23.480000    22.760000     9.232240
        8       142.800000   141.200000    54.408800    23.660000    23.480000     8.961520
        9       145.500000   142.800000    56.107200    24.100000    23.660000     9.288840
       10       145.300000   145.500000    54.217000    24.010000    24.100000     8.923400
       11       148.300000   145.300000    57.342200    24.540000    24.010000     9.509740
       12       146.400000   148.300000    53.564200    24.300000    24.540000     8.937960
       13       150.200000   146.400000    58.553600    25.000000    24.300000     9.788200
       14       153.100000   150.200000    59.074800    25.640000    25.000000     9.990000
       15       157.300000   153.100000    61.459400    26.360000    25.640000    10.309360
       16       160.700000   157.300000    62.230200    26.980000    26.360000    10.478640
       17       164.200000   160.700000    63.601800    27.520000    26.980000    10.630520
       18       165.600000   164.200000    62.810800    27.780000    27.520000    10.552480
       19       168.700000   165.600000    65.034400    28.240000    27.780000    10.849720
       20       171.700000   168.700000    66.093800    28.780000    28.240000    11.101760

>mglh

>model yp = constant + xp

>estimate
1 case(s) deleted due to missing data.
 
Dep Var: YP   N: 19   Multiple R: 0.997601   Squared multiple R: 0.995208
 
Adjusted squared multiple R: 0.994926   Standard error of estimate: 0.067195
 
Effect         Coefficient    Std Error     Std Coef Tolerance     t   P(2 Tail)
 
CONSTANT         -0.403282     0.167675     0.0        .       -2.40514  0.02784
XP                0.173821     0.002925     0.997601  1.000000 59.41838  0.00000
 
                             Analysis of Variance
 
Source             Sum-of-Squares   df  Mean-Square     F-ratio       P
 
Regression             15.941142     1    15.941142 3530.543519    0.000000
Residual                0.076759    17     0.004515
----------------------------------------------------------------------------------------------------------------------------------
 
 
Durbin-Watson D Statistic     1.644
First Order Autocorrelation   0.150

Now redo the D-W test; if D-W still significant one can iterate
the Cochrane-Orcutt procedure (see NKNW p. 510)

>rem Hildreth-Lu procedure using nonlin

>nonlin

>model y = rho*y1 + b0 + b1*(x-rho*x1)

>estimate
 
 Iteration
 No.      Loss      RHO         B0          B1
   0 .631740D+02 .101000D+01-.102000D+01 .103000D+01
   1 .133054D+00 .101044D+01 .848852D-01 .158714D+00
   2 .722882D-01 .967085D+00 .768116D-01 .159566D+00
   3 .716819D-01 .961296D+00 .718397D-01 .160309D+00
   4 .716707D-01 .959414D+00 .716441D-01 .160463D+00
   5 .716704D-01 .958978D+00 .716043D-01 .160508D+00
   6 .716704D-01 .958861D+00 .716071D-01 .160519D+00
   7 .716704D-01 .958831D+00 .716075D-01 .160522D+00
 
Dependent variable is Y
 
Zero weights, missing data or estimates reduced degrees of freedom
    Source   Sum-of-Squares    df  Mean-Square
 Regression     1.17437E+04     3  3914.570710
   Residual        0.071670    16     0.004479
 
      Total     1.17438E+04    19
Mean corrected    96.679779    18
 
       Raw  R-square (1-Residual/Total)        =     0.999994
Mean corrected R-square (1-Residual/Corrected) =     0.999259
          R(observed vs predicted) square      =     0.999263
 
                                                      Wald Confidence Interval
Parameter         Estimate       A.S.E.    Param/ASE        Lower < 95%> Upper
 RHO              0.958831     0.080054    11.977261     0.789123     1.128538
 B0               0.071607     0.071555     1.000733    -0.080082     0.223297
 B1               0.160522     0.007931    20.240415     0.143710     0.177335

>rem compare with NKNW Table 12.5 p. 513



>rem first differences procedure

>mglh

>let dy = y - y1

>let dx = x - x1

>list x x1 dx y y1 dy
  Case number            X           X1           DX            Y           Y1           DY
        1       127.300000      .            .          20.960000      .            .
        2       130.000000   127.300000     2.700000    21.400000    20.960000     0.440000
        3       132.700000   130.000000     2.700000    21.960000    21.400000     0.560000
        4       129.400000   132.700000    -3.300000    21.520000    21.960000    -0.440000
        5       135.000000   129.400000     5.600000    22.390000    21.520000     0.870000
        6       137.100000   135.000000     2.100000    22.760000    22.390000     0.370000
        7       141.200000   137.100000     4.100000    23.480000    22.760000     0.720000
        8       142.800000   141.200000     1.600000    23.660000    23.480000     0.180000
        9       145.500000   142.800000     2.700000    24.100000    23.660000     0.440000
       10       145.300000   145.500000    -0.200000    24.010000    24.100000    -0.090000
       11       148.300000   145.300000     3.000000    24.540000    24.010000     0.530000
       12       146.400000   148.300000    -1.900000    24.300000    24.540000    -0.240000
       13       150.200000   146.400000     3.800000    25.000000    24.300000     0.700000
       14       153.100000   150.200000     2.900000    25.640000    25.000000     0.640000
       15       157.300000   153.100000     4.200000    26.360000    25.640000     0.720000
       16       160.700000   157.300000     3.400000    26.980000    26.360000     0.620000
       17       164.200000   160.700000     3.500000    27.520000    26.980000     0.540000
       18       165.600000   164.200000     1.400000    27.780000    27.520000     0.260000
       19       168.700000   165.600000     3.100000    28.240000    27.780000     0.460000
       20       171.700000   168.700000     3.000000    28.780000    28.240000     0.540000
The first differences model is run without a constant term
>model dy = dx

>estimate
1 case(s) deleted due to missing data.
Model contains no constant
 
Dep Var: DY   N: 19   Multiple R: 0.991867   Squared multiple R: 0.983800
 
Adjusted squared multiple R: 0.983800   Standard error of estimate: 0.069392
 
Effect         Coefficient    Std Error     Std Coef Tolerance     t   P(2 Tail)
 
DX                0.168488     0.005096     0.991867  1.000000 33.06272  0.00000
 
                             Analysis of Variance
 
Source             Sum-of-Squares   df  Mean-Square     F-ratio       P
 
Regression              5.263726     1     5.263726 1093.143324    0.000000
Residual                0.086674    18     0.004815
----------------------------------------------------------------------------------------------------------------------------------
 
 
Durbin-Watson D Statistic     1.739
First Order Autocorrelation   0.122
But to calculate the D-W statistic one must run the same model WITH a
constant
>model dy = constant + dx

>estimate
1 case(s) deleted due to missing data.
 
Dep Var: DY   N: 19   Multiple R: 0.982747   Squared multiple R: 0.965791
 
Adjusted squared multiple R: 0.963778   Standard error of estimate: 0.065498
 
Effect         Coefficient    Std Error     Std Coef Tolerance     t   P(2 Tail)
 
CONSTANT          0.040528     0.022642     0.0        .        1.78996  0.09129
DX                0.158783     0.007248     0.982747  1.000000 21.90756  0.00000
 
                             Analysis of Variance
 
Source             Sum-of-Squares   df  Mean-Square     F-ratio       P
 
Regression              2.058923     1     2.058923  479.941009    0.000000
Residual                0.072929    17     0.004290
----------------------------------------------------------------------------------------------------------------------------------
 
*** WARNING ***
Case            4 has large leverage   (Leverage =     0.441713)
 
Durbin-Watson D Statistic     1.749
First Order Autocorrelation   0.116



Last modified 22 April 1999