### Present Remotely

Send the link below via email or IM

• Invited audience members will follow you as you navigate and present
• People invited to a presentation do not need a Prezi account
• This link expires 10 minutes after you close the presentation

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

# Ridge Regression & Nonlinear Regression

No description
by

## 蓁宜 劉

on 23 May 2014

Report abuse

#### Transcript of Ridge Regression & Nonlinear Regression

Ridge Regression &
Nonlinear Regression

1. 當自變數間有共線性

2. 當自變數多於資料量

X=[x1 x2 x3];
b=ridge(y,X,100)
plot(y);hold on;
y1=b(1)*x1+b(2)*x2+b(3)*x3;
plot(y1)

b =
1.2971
0.4976
-1.2408

x=4:0.1:12;
y=8*sin(x).*exp(x)+12.*log(x);
a=[0 1];
f=@(a,x)a(1)*sin(x).*exp(2*x)+a(2).*log(x);
a=nlinfit(x,y,f,a)
plot (x,y);hold on;
y1=a(1)*sin(x).*exp(2*x)+a(2).*log(x);
plot(x,y1);

a =

1.0e+004 *

0.0000 -1.5576
matlab tool中的nlinfit使用方法
Outline
Ridge Regression

Ridge Regression
Nonlinear Regression
Nonlinear Regression

matlab tool 中的nlinfit使用方法

Ridge regression，又稱Tikhonov regularization，中文翻譯為嶺回歸或脊回歸，是一種專用於共線性數據分析的有偏估計回歸方法，實質上是一種改良的最小二乘估計法，通過放棄最小二乘法的無偏性，以損失部分信息、降低精度為代價獲得回歸係數更為符合實際、更可靠的回歸方法，對病態數據的擬合要強於最小二乘法。

Full transcript