请大家帮忙看下这个分析
我有些问题,请知道的帮忙解释下啊:
The regression equation is
C1= 2.43 + 0.0803 C2
Predictor Coef SE Coef T P
Constant 2.4251 0.1917 12.65 0.000
C2 0.080281 0.005642 14.23 0.000
S = 0.813231 R-Sq = 80.2% R-Sq(adj) = 79.8%
Analysis of Variance
Source DF SS MS F P
Regression 1 133.89 133.89 202.45 0.000
Residual Error 50 33.07 0.66
Total 51 166.96
Unusual Observations
Obs C2 C1 Fit SE Fit Residual St Resid
23 22.3 6.193 4.217 0.116 1.976 2.46R
28 72.8 7.084 8.273 0.280 -1.189 -1.56 X
45 28.2 6.557 4.689 0.113 1.868 2.32R
48 15.6 5.469 3.678 0.131 1.791 2.23R
R denotes an observation with a large standardized residual.
X denotes an observation whose X value gives it large influence.
Predicted Values for New Observations
New
Obs Fit SE Fit 99% CI 99% PI
1 8.687 0.307 (7.866, 9.508) (6.360, 11.014)X
X denotes a point that is an outlier in the predictors.
Values of Predictors for New Observations
New
Obs C2
1 78.0
1)请问 R-sq R-sq(adj)要多少才能接受,如果很小,比如说10%,是不要去掉一些点再重新拟合,请问去掉的点的标准是什么,谢谢
2)78是C2的已知的规格值,请问从Predicted Values for New Observations开始是MTB中选哪个选项产生的,78在哪里输入,谢谢(找了很多选项都试不出(?
3)C1和C2是不都要正态分布做回归分析才有意义,如果P<0.05回归分析还有意义么
谢谢大家啦:D
The regression equation is
C1= 2.43 + 0.0803 C2
Predictor Coef SE Coef T P
Constant 2.4251 0.1917 12.65 0.000
C2 0.080281 0.005642 14.23 0.000
S = 0.813231 R-Sq = 80.2% R-Sq(adj) = 79.8%
Analysis of Variance
Source DF SS MS F P
Regression 1 133.89 133.89 202.45 0.000
Residual Error 50 33.07 0.66
Total 51 166.96
Unusual Observations
Obs C2 C1 Fit SE Fit Residual St Resid
23 22.3 6.193 4.217 0.116 1.976 2.46R
28 72.8 7.084 8.273 0.280 -1.189 -1.56 X
45 28.2 6.557 4.689 0.113 1.868 2.32R
48 15.6 5.469 3.678 0.131 1.791 2.23R
R denotes an observation with a large standardized residual.
X denotes an observation whose X value gives it large influence.
Predicted Values for New Observations
New
Obs Fit SE Fit 99% CI 99% PI
1 8.687 0.307 (7.866, 9.508) (6.360, 11.014)X
X denotes a point that is an outlier in the predictors.
Values of Predictors for New Observations
New
Obs C2
1 78.0
1)请问 R-sq R-sq(adj)要多少才能接受,如果很小,比如说10%,是不要去掉一些点再重新拟合,请问去掉的点的标准是什么,谢谢
2)78是C2的已知的规格值,请问从Predicted Values for New Observations开始是MTB中选哪个选项产生的,78在哪里输入,谢谢(找了很多选项都试不出(?
3)C1和C2是不都要正态分布做回归分析才有意义,如果P<0.05回归分析还有意义么
谢谢大家啦:D
没有找到相关结果
已邀请:
3 个回复
Jeff_wang (威望:3) (江苏 )
赞同来自:
- R-sq和R-sq(adj)实际体现了当前的模型在多大程度上能解释总变差。因此这两个值肯定高了好,10%意味着模型说明不了流程,其它因子远比当前模型中的影响显著。
调整模型也需分析R-sq和R-sq(adj)低的原因。如果某关键因子根本没考虑,在当前模型上增减调整就起不到作用。如果因子都包考虑到了,可使用best subset方法帮助筛选优化模型,也许还会结合fitted line plot看有无高次项影响。另加一句:从模型中剔除某项,一定先去掉那些不显著的因子(项)。