| 查看: 815 | 回复: 6 | ||
| 【奖励】 本帖被评价4次，作者alwens增加金币 3.5 个 | ||
| 当前主题已经存档。 | ||
| 当前只显示满足指定条件的回帖，点击这里查看本话题的所有回帖 | ||
[资源]
关于QSAR模型的评价
|
||
|
在QSAR中一般对线性回归模型的回归或预测性能会用相关系数R进行评价，对于初学者应注意评价模型优劣的用词，一般4个常用的描述性词汇（poor，fair，good，excellent）有对应的量化指标，如使用不当容易引起审稿人的反感。 ps: 除了相关系数，还应该同时考虑所得到模型的截距（应接近0）和斜率（应接近1），有文章使用下表中的标准，仅供参考。 Ranking of results according to the correlation indexes： Index Excellent Good Fair Poor ----------------------------------------------------------------------------- R 1.00-0.90 0.89-0.80 0.79-0.50 <0.50 Slope 1.00-0.90 0.89-0.80 0.79-0.50 <0.50 Intercept 0.00-0.10 0.11-0.20 0.20-0.50 >0.50 [ Last edited by alwens on 2008-1-17 at 16:30 ] |
回复此楼» 猜你喜欢
B区学生调剂-兰州交通大学材料科学与工程学院
已经有12人回复
-
山西大同大学物理学专业还有调剂名额，欢迎调剂！
已经有13人回复
-
物理学I论文润色/翻译怎么收费?
已经有79人回复
-
桂林理工大学物理学专业招收调剂
已经有18人回复
-
VASP 的一组 GPU / CPU 基准测试记录
已经有0人回复
-
津理工大学晶体材料全国重点实验室刘红军教授课题组招收博士生一名
已经有0人回复
-
【原创讨论】从电子约束到物质编辑：一套可迭代的环形磁场科技树
已经有0人回复
-
【方案分享】单环磁场+轴心控制+偏转导出电子束约束系统（可行性实验）
已经有6人回复
-
【修正版】单环用磁约束低速电子实验方案（简化版）
已经有0人回复
-
桂林理工大学物理学专业招收调剂，还有三个名额！！！
已经有18人回复
3楼2008-02-24 22:22:47
★★★★★ 五星级,优秀推荐
★
lei0736(金币+1,VIP+0):多谢积极参与讨论
lei0736(金币+1,VIP+0):多谢积极参与讨论
|
好像回归性是用R来评价的 预测性还不行吧， 除了R和交差验证R，还有一些其它指标。 下面是qsar鼻祖Hansch做的QSAR后，对其方程进行评价的内容： Validation of the QSAR Models QSAR model validation is an essential task in developing a statistically valid and predictive model, because the real utility of a QSAR model is in its ability to predict accurately the modeled property for new compounds. The following approaches have been used for the validation of QSAR Eqs. 1–20: • Fraction of the variance: The fraction of the variance of an MRA model is expressed by r2. It is believed that the closer the value of r2 to unity, the better the QSAR model. The values of r2 for these QSAR models are from 0.787 to 0.993, which suggests that these QSARmodels explain 78.7–99.3% of the variance of the data. According to the literature, the predictive QSAR model must have r2 > 0.6 [73, 74]. • Cross-validation test: The values of q2 for these QSAR models are from 0.549 to 0.972. The high values of q2 validate the QSAR models. From the literature, it must be greater than 0.50 [73, 74]. • Standard deviation (s): s is the standard deviation about the regression line. The smaller the value of s the better the QSAR model. The values of s for these QSAR models are from 0.065 to 0.406. • Quality factor or quality ratio (Q): The high values of Q (2.259–14.646) for these QSAR models suggest that the high predictive power for these models as well as no over-fitting. • Fischer statistics (F): Fischer statistics (F) is the ratio between explained and unexplained variance for a given number of degree of freedom. The larger the F value the greater the probability that the QSAR equation is significant. The F values obtained for these QSAR models are from 17.622 to 283.714, which are statistically significant at the 95% level. • All the QSAR models (except Eqs. 7 and 9) also fulfill the thumb rule condition that (number of data points)/(number of descriptors) ≥ 4. |
2楼2008-02-01 22:00:59
5