全部版块 我的主页
论坛 数据科学与人工智能 数据分析与数据科学 MATLAB等数学软件专版
3734 3
2010-01-22
     x1     x2    x3      y
-0.41-0.54-0.740.09
1.920.810.030.12
4.743.682.420.15
65.0340.2
43.593.130.17
0.160.420.590.16
0.32-0.21-0.590.1
2.231.020.090.14
6.214.462.620.2
6.415.434.330.18
4.273.83.30.18
-0.080.210.440.11
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

全部回复
2010-1-23 03:17:39
First of all, you should have a model: for example: Y = x1*exp(-a1)+x2^a2+x3*a3;

Second of all, there are bunch of matlab functions that you can use:
1). Minimize the squared errors of (Y-x1*exp(-a1)-x2^2-x3) -- fmincon or fminsearch if you have constraint.
2). Minimize the same statistics aforementioned by lsqnonlin or lsqnonneg, depending on the coefficients that you want to estimate.

Hopefully this will help.

By the way, there are many examples on the website of mathworks.com so that you would walk through to get a better idea of what kind of question that you are estimating and what methods you need to apply.

Good luck.
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

2010-1-24 17:00:29
工具箱里有吧
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

2010-1-25 16:03:36
谢谢楼上二位!
二维码

扫码加我 拉你入群

请注明:姓名-公司-职位

以便审核进群资格,未注明则拒绝

相关推荐
栏目导航
热门文章
推荐文章

说点什么

分享

扫码加好友,拉您进群
各岗位、行业、专业交流群