作者glasslands (剪刀)
看板Statistics
标题Re: [问题] 统计的假设检定问题(型I误差)
时间Fri Feb 15 10:00:26 2008
※ 引述《liliya ()》之铭言:
: A random sample of 400 voters in a certain city are asked if they
: favor an additional 4% gasoline sales tax to provide badly needed
: revenues for street repairs. If more than 220 but fewer than 260
: favor the sales tax, we shall conclude that 60% of the voters
: are for it.
: (1) If 60% of the voters favor the incresed tax, what is the type I
: error?
题目应是问说 估母体的比例
现在有样本400 并得到一个样本比例p*
(依中央及限定理 抽样分配是个常态分配)
P(220/400<p<260/400)=1-阿发
标准化:
P((55%-60%/标准差 < z < 65%-60%/标准差)=1-type1 error
标准差是抽样分配的标准差 也就是(60%)(1-60%)/400的更号
这样应该就可算出type1了
(2)If actually only 48% of the voters are in favor of the
: additional gasoline tax, what is the typeII error?
P(55%-48%/标准差 < z < 65%-48%/标准差)=2
标准差要用0.48算! (严格说来这是标准误)
简单来说~~就是当母体比例是0.48
样本比例却是0.55-0.65之间 然後被接受了! 这样的机率是多少
这是我的想法^^ 你可以参考看看
: 我知道 型I误差 =P(拒绝H0|H0为真)
: 但搞不太懂他的H0 欲否定的假设是??? Orz
: 请问大家 大概的意思 假设是????
: 谢谢大家的帮忙!! ^^"
--
※ 发信站: 批踢踢实业坊(ptt.cc)
◆ From: 59.113.171.119
1F:推 MIZUYAMA:阿发XDDDD 有本土味 02/16 22:52
2F:推 liliya:算完typeI error 可是没有这选项Orz 02/18 16:22
3F:→ glasslands:你算TYPE1 ERROR是0.0414吗? 02/18 23:10
4F:推 liliya:恩 对 我是算0.0414没错 但没这选项>"< 02/19 13:26