sas中可以实现检验某个分布是否属于t分布的功能吗？

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

   sas中可以实现检验某个分布是否属于t分布的功能吗？对数据做了正态性检验，发现不满足，但是看图觉得有尖峰肥尾的特性，想检验一下数据是否符合t分布？自由度是多少？谢谢啦~~~

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

分享0 收藏0 回帖

关键词：t分布 正态性检验 是多少 自由度 检验

从t分布定义可知，前提条件随机变量应该服从正态分布，而不符合正态分布的变量还考虑服从t分布？

使用拟合优度检验  cha**2

老师她摸我 发表于 2012-11-21 22:12
从t分布定义可知，前提条件随机变量应该服从正态分布，而不符合正态分布的变量还考虑服从t分布？

啊……你说的是不是t检验……

是t分布，建议你看看数理统计的抽样分布

I think I understand your question, because I have this question as well.
Unfortunately, there is NO way in SAS for testing if data follow by T distribution. Usually we call T distribution a 'fatter' normal one. So, fatter tail is true but not sharp mode (peak). so based on your plot, it does not look like T-distributed at all.
where we apply T-test, we believe the test statistic follows by a T -distribution. It may virtually be normally-distributed but not normal because we do have normal test. For example, count data on binomial.
Jingju

jingju11 发表于 2012-11-22 10:15
I think I understand your question, because I have this question as well.
Unfortunately, there is N ...

Thank you very much! But I don't quite understand your saying of "counting data on binomial". Do you mean I can count the frequency of different intervals instead of fit a normal distribution?