参数和非参数方法之间的区别
参数方法
参数方法背后的基本思想是,有一组固定参数用于确定机器学习中使用的概率模型。
参数方法是那些我们事先知道总体是正态的方法,或者如果不是,那么我们可以使用正态分布轻松地近似它,这可以通过调用中心极限定理来实现。
使用正态分布的参数是 –
- 意思
- 标准差
最终,参数化方法的分类完全取决于对总体所做的假设。有许多可用的参数方法,其中一些是:
- 用于 – 总体均值和已知标准差的置信区间。
- 置信区间用于 – 总体均值以及未知标准差。
- 总体方差的置信区间。
- 两个均值之差的置信区间,标准差未知。
非参数方法
在非参数方法中,不需要对给定的总体或我们正在研究的总体进行任何参数假设。事实上,这些方法并不取决于人口。
这里没有可用的固定参数集,也没有任何类型的分布(正态分布等)可供使用。这也是非参数方法也称为无分布方法的原因。
如今,非参数方法越来越受欢迎,而这种名声背后的一些原因是——
- 主要的原因是使用参数化方法时不需要有礼貌。
- 第二个重要原因是,我们不需要对我们正在研究的给定(或采用)总体做出越来越多的假设。
- 大多数可用的非参数方法都非常容易应用和理解,即复杂性非常低。
今天有许多非参数方法可用,但其中一些是——
- 斯皮尔曼相关性检验
- 总体均值的符号检验
- 两个独立均值的 U 检验
参数和非参数方法之间的区别如下 -
S.No. | Parametric Methods | Non-Parametric Methods |
---|---|---|
1. | Parametric Methods uses a fixed number of parameters to build the model. | Non-Parametric Methods use the flexible number of parameters to build the model. |
2. | Parametric analysis is to test group means. | A non-parametric analysis is to test medians. |
3. | It is applicable only for variables. | It is applicable for both – Variable and Attribute. |
4. | It always considers strong assumptions about data. | It generally fewer assumptions about data. |
5. | Parametric Methods require lesser data than Non-Parametric Methods. | Non-Parametric Methods requires much more data than Parametric Methods. |
6. | Parametric methods assumed to be a normal distribution. | There is no assumed distribution in non-parametric methods. |
7. | Parametric data handles – Intervals data or ratio data. | But non-parametric methods handle original data. |
8. | Here when we use parametric methods then the result or outputs generated can be easily affected by outliers. | When we use non-parametric methods then the result or outputs generated cannot be seriously affected by outliers. |
9. | Parametric Methods can perform well in many situations but its performance is at peak (top) when the spread of each group is different. | Similarly, Non-Parametric Methods can perform well in many situations but its performance is at peak (top) when the spread of each group is the same. |
10. | Parametric methods have more statistical power than Non-Parametric methods. | Non-parametric methods have less statistical power than Parametric methods. |
11. | As far as the computation is considered these methods are computationally faster than the Non-Parametric methods. | As far as the computation is considered these methods are computationally faster than the Parametric methods. |
12. | Examples – Logistic Regression, Naïve Bayes Model, etc. | Examples – KNN, Decision Tree Model, etc. |