由买买提看人间百态

boards

本页内容为未名空间相应帖子的节选和存档,一周内的贴子最多显示50字,超过一周显示500字 访问原贴
Statistics版 - 关于函数的逼近问题
相关主题
data prediction by regression or better ways如何根据cdf(比较复杂,不能求微分)求其pdf
How to compare two correlations in SAS?统计经典教科书
请推荐smoothing&nonparametric regression方面的参考书Re: 请推荐nonparametric regression 的入门经典书
请教一个data analysis question现代统计学大家: Eric Lehmann 【ZZ】
问一个求2个变量correlation的问题抛砖引玉:敢问路在何方?
Predict values of vectors generated by black box functions请问Methods of Nonparametric Statistics这门课重要吗?
关于基因表达的统计学问题, 不知发这是否合适求教一个sas或STATA 画95%CI的问题!
怎样在data step跳过格式不对的data record?有人来谈谈统计学的未来么?
相关话题的讨论汇总
话题: m2话题: m1话题: any话题: find
进入Statistics版参与讨论
1 (共1页)
l******9
发帖数: 579
1
I have two functions
m1 = f1(w, s)
m2 = f2(w, s)
f1() and f2() are all blackboxs. Given w and s, I can get m1 and m2.
Now, I need to design or find a function g, such that
m2' = g(m1)
Also, the difference between m2 and m2' must be minimized.
The w and s are all stochastic process.
How can I find such a function g()? What knowledge domain does this belong
to ?
Any help would be appreciated !
v*******e
发帖数: 11604
2
打印个scatter plot, m2 vs m1;就差不多知道了。看起来像线性关系就做线性回归,
想别的关系就再说。
l******9
发帖数: 579
3
m1 and m2 have no obvious linear relationships !
Any help would be appreciated !

【在 v*******e 的大作中提到】
: 打印个scatter plot, m2 vs m1;就差不多知道了。看起来像线性关系就做线性回归,
: 想别的关系就再说。

c********h
发帖数: 330
4
nonparametric methods for arbitrary relationship might be helpful.
Try smoothing splines on y = m2, x = m1, though it may be overkill~
c*****l
发帖数: 1493
5
这问题和f1f2完全没关系,
正如楼上所说,就是个regression。
定义你的distance,找对应的projection matrix。

【在 l******9 的大作中提到】
: I have two functions
: m1 = f1(w, s)
: m2 = f2(w, s)
: f1() and f2() are all blackboxs. Given w and s, I can get m1 and m2.
: Now, I need to design or find a function g, such that
: m2' = g(m1)
: Also, the difference between m2 and m2' must be minimized.
: The w and s are all stochastic process.
: How can I find such a function g()? What knowledge domain does this belong
: to ?

c***z
发帖数: 6348
6
m2' = g(m1)
solving ODE numerically?
not an expert but you can find help at math board

【在 l******9 的大作中提到】
: I have two functions
: m1 = f1(w, s)
: m2 = f2(w, s)
: f1() and f2() are all blackboxs. Given w and s, I can get m1 and m2.
: Now, I need to design or find a function g, such that
: m2' = g(m1)
: Also, the difference between m2 and m2' must be minimized.
: The w and s are all stochastic process.
: How can I find such a function g()? What knowledge domain does this belong
: to ?

c********h
发帖数: 330
7
我猜楼主的那个'不是求导的意思?

【在 c***z 的大作中提到】
: m2' = g(m1)
: solving ODE numerically?
: not an expert but you can find help at math board

l******9
发帖数: 579
8
@catforfish, You are right.
Any help would be appreciated.

【在 c********h 的大作中提到】
: 我猜楼主的那个'不是求导的意思?
1 (共1页)
进入Statistics版参与讨论
相关主题
有人来谈谈统计学的未来么?问一个求2个变量correlation的问题
对于 interval censor 的情况怎么写SAS code呀?Predict values of vectors generated by black box functions
[合集] 问个简单的问题关于基因表达的统计学问题, 不知发这是否合适
有没有必要从生统转到统计?怎样在data step跳过格式不对的data record?
data prediction by regression or better ways如何根据cdf(比较复杂,不能求微分)求其pdf
How to compare two correlations in SAS?统计经典教科书
请推荐smoothing&nonparametric regression方面的参考书Re: 请推荐nonparametric regression 的入门经典书
请教一个data analysis question现代统计学大家: Eric Lehmann 【ZZ】
相关话题的讨论汇总
话题: m2话题: m1话题: any话题: find