[討論] What is p-value NOT

看板Statistics作者 (偶素米蟲)時間12年前 (2011/09/04 12:54), 編輯推噓8(8推 0噓 23→)
留言31則, 6人參與, 最新討論串1/1
p-value是什麼應該是FAQ了, 來篇p-value不是什麼好了 - The p-value is not the probability that H0 is true. - The p-value is not the probability that H0 is false. - The p-value is not the probability that H1 is true. - The p-value is not the probability that H1 is false. - The p-value is not the probability that your results are due to random chance. - The p-value is not the probability that your results are not due to random chance. There is one other thing: the p-value is NOT a measure of the size or importance of an effect. A small p-value does not tell you by how much you are rejecting H0 (the effect size), or whether that rejection has any practical consequences. 引自 http://www.tc3.edu/instruct/sbrown/stat/pvalue.htm -- ※ 發信站: 批踢踢實業坊(ptt.cc) ◆ From: 173.73.120.121
推
09/04 16:52, , 1F
一直以為P-value是第5項 假定P-value=0.01
09/04 16:52, 1F
→
09/04 16:53, , 2F
難道不是在說在100次取樣的 test statistic 中
09/04 16:53, 2F
→
09/04 16:55, , 3F
會有一次隨機發生如手上P=0.01的樣本的 test statistic 嗎
09/04 16:55, 3F
→
09/04 16:56, , 4F
(假定不連續)
09/04 16:56, 4F
推
09/04 18:03, , 5F
p-value是一個conditional probability，所以重點是那個
09/04 18:03, 5F
→
09/04 18:03, , 6F
condition是什麼
09/04 18:03, 6F
推
09/04 20:49, , 7F
看不懂最後一段，如果全部都是否定的p-value到底是什麼
09/04 20:49, 7F
→
09/04 22:19, , 8F
wertymin24,貼這篇文章的主要目的正式要讓大家思考p-value
09/04 22:19, 8F
→
09/04 22:21, , 9F
到底是什麼(可以吃嗎),很多人會背定義,但是沒有真正了解定義
09/04 22:21, 9F
→
09/04 22:21, , 10F
你認知的p-value是什麼?
09/04 22:21, 10F
→
09/04 22:28, , 11F
g大, 你可以從p-value判斷這個研究資料是否是隨機取樣嗎?
09/04 22:28, 11F
→
09/04 22:33, , 12F
反過來說,如果研究設計是隨機取樣,那不管觀察值是什麼
09/04 22:33, 12F
→
09/04 22:34, , 13F
都是隨機取樣而來的(observed data are all due to random
09/04 22:34, 13F
→
09/04 22:34, , 14F
chance)
09/04 22:34, 14F
推
09/05 01:31, , 15F
我認知的定義比較操作性一點，以下文字取自wiki
09/05 01:31, 15F
→
09/05 01:31, , 16F
One often "rejects the null hypothesis" when the
09/05 01:31, 16F
→
09/05 01:31, , 17F
p-value is less than the significance level α
09/05 01:31, 17F
→
09/05 01:32, , 18F
, which is often 0.05 or 0.01.
09/05 01:32, 18F
→
09/05 01:32, , 19F
When the null hypothesis is rejected, the result i
09/05 01:32, 19F
→
09/05 01:32, , 20F
is said to be statistically significant.
09/05 01:32, 20F
→
09/05 02:36, , 21F
所以呢?
09/05 02:36, 21F
→
09/05 02:44, , 22F
跟最後一段有衝突嗎?
09/05 02:44, 22F
推
09/05 03:13, , 23F
P-value 我後來都用 false positive 的觀點去想
09/05 03:13, 23F
→
09/05 03:17, , 24F
好像用 AUC of ROC 會比直接看 P-value 來的有用
09/05 03:17, 24F
→
09/05 03:17, , 25F
因為 type I error 只講了一半的事情
09/05 03:17, 25F
推
09/05 10:49, , 26F
bmka大~我知道沒有衝突啦~我只是回應你，你問的我認知
09/05 10:49, 26F
→
09/05 10:50, , 27F
的pvalue是什麼
09/05 10:50, 27F
推
09/08 00:36, , 28F
應該說 不要太執著於做出有很多星星的文章
09/08 00:36, 28F
→
09/08 00:36, , 29F
而是給予目前這個結果的合理解釋或者說法
09/08 00:36, 29F
→
09/08 01:52, , 30F
是的,貼這篇正是為了回應上一篇文章
09/08 01:52, 30F
推
09/08 19:29, , 31F
一顆星 兩顆星 三顆星 XD
09/08 19:29, 31F
‣ 更多 bmka 的文章
文章代碼(AID): #1EOmIDv9 (Statistics)