环境风险评价作业

1. 根据数据homework，利用SPSS软件，建立因变量y2与自变量 x1，x2，x3，x4和y1之间的回归方程，并对分析结果进行简要解释。

第一种方法： 非线性回归.

MODEL PROGRAM a=1 b=1 c=1 d=1 e=1 f=1.

假设方程为y1=a * x1+b / x2+c / x3+d * EXP(e * x4) + f * y2.

迭代历史记录 迭代数 ab残差平方和 a b 1.000 147.067 147.067 -2047.506 -1685.260 -1685.260 -1664.754 -1664.754 -1655.506 -1655.506 -1652.696 -1652.696 -1648.135 -1637.155 -1637.155 -1637.950 -1637.950 -1634.615 -1634.615 -1635.876 -1635.876 -1634.556 -1634.556 -1634.088 -1634.088 -1633.884 -1633.884 -1633.995 -1633.995 -1634.564 -1634.564 c 参数 d 1.000 -2.936E-007 -2.936E-007 5.549E-007 6.286E-008 6.286E-008 2.852E-008 2.852E-008 2.792E-008 2.792E-008 5.631E-008 5.631E-008 2.409E-007 8.296E-008 8.296E-008 1.559E-007 1.559E-007 2.596E-007 2.596E-007 4.784E-007 4.784E-007 9.333E-007 9.333E-007 1.872E-006 1.872E-006 3.833E-006 3.833E-006 7.922E-006 7.922E-006 1.642E-005 1.642E-005 e 1.000 1.000 1.000 1.080 1.008 1.008 .992 .992 .958 .958 .890 .890 .754 .873 .873 .840 .840 .825 .825 .795 .795 .766 .766 .737 .737 .707 .707 .677 .677 .647 .647 f 1.000 -4.922 -4.922 -3.543 -3.447 -3.447 -3.442 -3.442 -3.438 -3.438 -3.435 -3.435 -3.429 -3.426 -3.426 -3.425 -3.425 -3.422 -3.422 -3.421 -3.421 -3.418 -3.418 -3.416 -3.416 -3.414 -3.414 -3.412 -3.412 -3.410 -3.410 1.0 1.1 2.0 2.1 2.2 3.0 3.1 4.0 4.1 5.0 5.1 6.0 6.1 6.2 7.0 7.1 8.0 8.1 9.0 9.1 10.0 10.1 11.0 11.1 12.0 12.1 13.0 13.1 14.0 14.1 15.0 339901005035634100000.000 31850238.404 31850238.404 4456870006.787 3424231.690 3424231.690 1754668.937 1754668.937 1641544.356 1641544.356 1628701.163 1628701.163 1633311.562 1628531.810 1628531.810 1628499.192 1628499.192 1627991.805 1627991.805 1627874.126 1627874.126 1627564.564 1627564.564 1627215.591 1627215.591 1626820.234 1626820.234 1626366.542 1626366.542 1625875.575 1625875.575 1.000 30.963 30.963 46.825 46.486 46.486 46.467 46.467 46.458 46.458 46.454 46.454 46.446 46.436 46.436 46.436 46.436 46.431 46.431 46.431 46.431 46.428 46.428 46.425 46.425 46.422 46.422 46.418 46.418 46.415 46.415 1.000 14271.473 14271.473 7880.723 7394.520 7394.520 7365.097 7365.097 7356.559 7356.559 7360.913 7360.913 7372.397 7357.982 7357.982 7363.590 7363.590 7368.679 7368.679 7374.759 7374.759 7382.181 7382.181 7390.559 7390.559 7399.951 7399.951 7410.339 7410.339 7421.736 7421.736 15.1 16.0 16.1 17.0 17.1 18.0 18.1 19.0 19.1 20.0 20.1 21.0 21.1 22.0 22.1 23.0 23.1 24.0 24.1 25.0 25.1 25.2 26.0 26.1 26.2 27.0 27.1 28.0 28.1 29.0 29.1 29.2 30.0 30.1 30.2

1625356.899 1625356.899 1624816.652 1624816.652 1624261.316 1624261.316 1623699.872 1623699.872 1623134.972 1623134.972 1622550.759 1622550.759 1622139.282 1622139.282 1621791.385 1621791.385 1621600.421 1621600.421 1620880.768 1620880.768 1621592.630 1620680.776 1620680.776 1620716.083 1620674.432 1620674.432 1620673.576 1620673.576 1620671.724 1620671.724 1620672.149 1620671.412 1620671.412 1620671.517 1620671.409 46.410 46.410 46.405 46.405 46.399 46.399 46.392 46.392 46.382 46.382 46.369 46.369 46.353 46.353 46.331 46.331 46.302 46.302 46.277 46.277 46.307 46.279 46.279 46.295 46.283 46.283 46.290 46.290 46.287 46.287 46.290 46.288 46.288 46.287 46.288 -1635.623 -1635.623 -1637.222 -1637.222 -1639.409 -1639.409 -1642.232 -1642.232 -1645.702 -1645.702 -1649.721 -1649.721 -1654.881 -1654.881 -1660.716 -1660.716 -1667.623 -1667.623 -1672.445 -1672.445 -1665.707 -1672.064 -1672.064 -1668.093 -1670.872 -1670.872 -1669.094 -1669.094 -1669.948 -1669.948 -1669.079 -1669.623 -1669.623 -1669.882 -1669.662 7434.215 7434.215 7447.867 7447.867 7462.802 7462.802 7479.130 7479.130 7496.886 7496.886 7515.777 7515.777 7536.360 7536.360 7558.445 7558.445 7581.481 7581.481 7596.565 7596.565 7579.807 7595.012 7595.012 7586.179 7592.749 7592.749 7588.711 7588.711 7590.771 7590.771 7588.757 7590.027 7590.027 7590.634 7590.120 3.404E-005 3.404E-005 7.053E-005 7.053E-005 .000 .000 .000 .000 .001 .001 .001 .001 .003 .003 .005 .005 .010 .010 .018 .018 .007 .017 .017 .012 .016 .016 .014 .014 .015 .015 .014 .014 .014 .015 .015 .617 .617 .587 .587 .556 .556 .526 .526 .496 .496 .468 .468 .438 .438 .408 .408 .377 .377 .362 .362 .393 .369 .369 .381 .372 .372 .377 .377 .375 .375 .377 .376 .376 .375 .376 -3.408 -3.408 -3.406 -3.406 -3.405 -3.405 -3.405 -3.405 -3.405 -3.405 -3.406 -3.406 -3.408 -3.408 -3.411 -3.411 -3.415 -3.415 -3.419 -3.419 -3.414 -3.419 -3.419 -3.416 -3.418 -3.418 -3.417 -3.417 -3.417 -3.417 -3.417 -3.417 -3.417 -3.417 -3.417 导数是通过数字计算的。

a. 主迭代数在小数左侧显示，次迭代数在小数右侧显示。

b. 由于连续残差平方和之间的相对减少量最多为 SSCON = 1.000E-008，因此在 66 模型评估和 30 导数评估之后，系统停止运行。

参数估计值 参数 估计 标准误 95% 置信区间 下限 a b c d e f 46.288 -1669.662 7590.120 .015 .376 -3.417 4.860 1561.474 3085.473 .226 .678 .792 36.696 -4751.654 1500.103 -.432 -.964 -4.981 上限 55.879 1412.330 13680.138 .461 1.715 -1.854 参数估计值的相关性 a b c d e f a 1.000 -.721 -.560 -.160 .157 .106 b -.721 1.000 .104 -.125 .129 .496 c -.560 .104 1.000 .156 -.152 -.107 d -.160 -.125 .156 1.000 -.999 -.199 e .157 .129 -.152 -.999 1.000 .207 f .106 .496 -.107 -.199 .207 1.000

ANOVA

源 回归 残差

未更正的总计 已更正的总计 因变量: y1

a. R 方 = 1 -（残差平方和）/（已更正的平方和）= .814。

平方和 55619049.284 1620671.409 57239720.693 8728643.213 df

6 173 179 178 均方 9269841.547 9368.043 a

Y1=46.288x1-1669.662/x2+7590.120/x3+0.015e0.376x4-3.417y2;R2=0.814

第二种方法：逐步线性回归

输入／移去的变量

模型

输入的变量

移去的变量

方法

a

步进（准则: F-to-enter 的1 x1 . 概率 <= .050，F-to-remove 的概率 >= .100）。 步进（准则: F-to-enter 的2 y2 . 概率 <= .050，F-to-remove 的概率 >= .100）。 步进（准则: F-to-enter 的3 x3 . 概率 <= .050，F-to-remove 的概率 >= .100）。 步进（准则: F-to-enter 的4 x2 . 概率 <= .050，F-to-remove 的概率 >= .100）。 a. 因变量: y1

模型汇总 模型 R R 方 调整 R 方 标准 估计的误差 1 2 3 4 .926 .931 .933 .936 dcba.857 .867 .871 .876 .857 .865 .869 .873 83.86486 81.27594 80.10710 78.83928 a. 预测变量: (常量), x1。 b. 预测变量: (常量), x1, y2。 c. 预测变量: (常量), x1, y2, x3。 d. 预测变量: (常量), x1, y2, x3, x2。 Anova 模型 回归 1 残差 总计 2 回归 平方和 7483746.608 1244896.606 8728643.213 7566026.079 df 1 177 178 2 均方 7483746.608 7033.314 F 1064.043 Sig. .000 ba 3783013.039 572.682 .000 c