４つの頑健推定の結果の比較

                                      Equation   1
                                      ============

                       Method of estimation = Ordinary Least Squares


 Dependent variable: PAI
 Current sample:  1953:1 to 1971:7
 Number of observations:  223

        Mean of dep. var. = 2.34560      LM het. test = 2.96299 [.085]
   Std. dev. of dep. var. = 3.24604     Durbin-Watson = 2.18459 [.906,.927]
 Sum of squared residuals = 1786.22  Jarque-Bera test = 4.43683 [.109]
    Variance of residuals = 8.08243   Ramsey's RESET2 = .127172 [.722]
 Std. error of regression = 2.84296   F (zero slopes) = 68.4129 [.000]
                R-squared = .236385    Schwarz B.I.C. = 553.827
       Adjusted R-squared = .232930    Log likelihood = -548.420

            Estimated    Standard
 Variable  Coefficient     Error       t-statistic   P-value
 C         -.867767      .431225       -2.01233      [.045]
 TB1       1.01470       .112201       9.04353       [.000]
 Standard Errors are heteroskedastic-consistent (HCTYPE=0).


                                      Equation   2
                                      ============

                       Method of estimation = Ordinary Least Squares


 Dependent variable: PAI
 Current sample:  1953:1 to 1971:7
 Number of observations:  223

        Mean of dep. var. = 2.34560      LM het. test = 2.96299 [.085]
   Std. dev. of dep. var. = 3.24604     Durbin-Watson = 2.18459 [.906,.927]
 Sum of squared residuals = 1786.22  Jarque-Bera test = 4.43683 [.109]
    Variance of residuals = 8.08243   Ramsey's RESET2 = .127172 [.722]
 Std. error of regression = 2.84296   F (zero slopes) = 68.4129 [.000]
                R-squared = .236385    Schwarz B.I.C. = 553.827
       Adjusted R-squared = .232930    Log likelihood = -548.420

            Estimated    Standard
 Variable  Coefficient     Error       t-statistic   P-value
 C         -.867767      .433171       -2.00329      [.046]
 TB1       1.01470       .112708       9.00289       [.000]
 Standard Errors are heteroskedastic-consistent (HCTYPE=1).


                                      Equation   3
                                      ============

                       Method of estimation = Ordinary Least Squares


 Dependent variable: PAI
 Current sample:  1953:1 to 1971:7
 Number of observations:  223

        Mean of dep. var. = 2.34560      LM het. test = 2.96299 [.085]
   Std. dev. of dep. var. = 3.24604     Durbin-Watson = 2.18459 [.906,.927]
 Sum of squared residuals = 1786.22  Jarque-Bera test = 4.43683 [.109]
    Variance of residuals = 8.08243   Ramsey's RESET2 = .127172 [.722]
 Std. error of regression = 2.84296   F (zero slopes) = 68.4129 [.000]
                R-squared = .236385    Schwarz B.I.C. = 553.827
       Adjusted R-squared = .232930    Log likelihood = -548.420

            Estimated    Standard
 Variable  Coefficient     Error       t-statistic   P-value
 C         -.867767      .433708       -2.00081      [.047]
 TB1       1.01470       .112991       8.98033       [.000]
 Standard Errors are heteroskedastic-consistent (HCTYPE=2).


                                      Equation   4
                                      ============

                       Method of estimation = Ordinary Least Squares


 Dependent variable: PAI
 Current sample:  1953:1 to 1971:7
 Number of observations:  223

        Mean of dep. var. = 2.34560      LM het. test = 2.96299 [.085]
   Std. dev. of dep. var. = 3.24604     Durbin-Watson = 2.18459 [.906,.927]
 Sum of squared residuals = 1786.22  Jarque-Bera test = 4.43683 [.109]
    Variance of residuals = 8.08243   Ramsey's RESET2 = .127172 [.722]
 Std. error of regression = 2.84296   F (zero slopes) = 68.4129 [.000]
                R-squared = .236385    Schwarz B.I.C. = 553.827
       Adjusted R-squared = .232930    Log likelihood = -548.420

            Estimated    Standard
 Variable  Coefficient     Error       t-statistic   P-value
 C         -.867767      .436212       -1.98932      [.048]
 TB1       1.01470       .113789       8.91738       [.000]
 Standard Errors are heteroskedastic-consistent (HCTYPE=3).