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 - A question about estimator
相关主题
A question about basic statistics, thanks.Google面试问题
Machine learning里的variance怎么算?Why shrinkage estimators are prefered?
请教: Var(Xi-X-bar) 等于Var(Xi) 吗?sample distribution的理解
有关EM algorithm 的问题请教一个bootstrapping的问题。
怎么分辨 standard error 和 standard deviation?问一个数据bias的问题
统计的核心课程(或者比较有用的课)有啥?请教个计算standard error的问题
问个binomial variance的公式Bias-variance 和Bayes factor 都可以用来比较不同的model, 但他们之间有何区别呢?谢谢。
统计牛人来帮个忙啊问一个pooled variance的问题。
相关话题的讨论汇总
话题: estimator话题: variance话题: bias话题: sample
进入Statistics版参与讨论
1 (共1页)
g**d
发帖数: 723
1
We know for an estimator that there is variance vs bias trade-off. It is
easy to understand the bias for an estimator. But what is the variance of an
estimator? It is definitely not the variance in the sample data, but
someone argue that since the estimator is deterministic, then there is no
variance in the estimator. Could someone help me out here please? Thanks.
D******n
发帖数: 2836
2
u messed up estimate with estimator
estimator is a function on sample,
estimate is a value.

an

【在 g**d 的大作中提到】
: We know for an estimator that there is variance vs bias trade-off. It is
: easy to understand the bias for an estimator. But what is the variance of an
: estimator? It is definitely not the variance in the sample data, but
: someone argue that since the estimator is deterministic, then there is no
: variance in the estimator. Could someone help me out here please? Thanks.

g**d
发帖数: 723
3
Ok, I kind of disagree with you, but I don't exactly know why. :)

【在 D******n 的大作中提到】
: u messed up estimate with estimator
: estimator is a function on sample,
: estimate is a value.
:
: an

q**j
发帖数: 10612
4
not variance-bias trade off. should be MSE (mean square error) and bias trad
e off. an estimator is also a statistic, so it depends on sample. with diffe
rent sample, estimator has different value, so it has a variance.

an

【在 g**d 的大作中提到】
: We know for an estimator that there is variance vs bias trade-off. It is
: easy to understand the bias for an estimator. But what is the variance of an
: estimator? It is definitely not the variance in the sample data, but
: someone argue that since the estimator is deterministic, then there is no
: variance in the estimator. Could someone help me out here please? Thanks.

g**d
发帖数: 723
D******n
发帖数: 2836
6
let d be an estimator ,S be sample
deterministic means Var_d(S) (d(S)) = 0;
variance means Var_S (d(S)) !=0 ;
it is very simple, from S to d(S) it is deterministic,not probabalistic,
variance of d(S) is in terms of S, S is random so d(S) is also random

【在 g**d 的大作中提到】
: http://www.aiaccess.net/English/Glossaries/GlosMod/e_gm_bias_variance.htm
: 查到这个, 但是还没看完.

1 (共1页)
进入Statistics版参与讨论
相关主题
问一个pooled variance的问题。怎么分辨 standard error 和 standard deviation?
问一个用simulation的问题统计的核心课程(或者比较有用的课)有啥?
2^2 factorial design without Replicates怎么ANOVA 分析啊?问个binomial variance的公式
统计面试,同胞何苦为难同胞。。。统计牛人来帮个忙啊
A question about basic statistics, thanks.Google面试问题
Machine learning里的variance怎么算?Why shrinkage estimators are prefered?
请教: Var(Xi-X-bar) 等于Var(Xi) 吗?sample distribution的理解
有关EM algorithm 的问题请教一个bootstrapping的问题。
相关话题的讨论汇总
话题: estimator话题: variance话题: bias话题: sample