开发者:上海品职教育科技有限公司 隐私政策详情

应用版本:4.2.11(IOS)|3.2.5(安卓)APP下载

沪上小王子 · 2024年02月05日

高频数据对变量间的异步更敏感,因此往往产生较低的相关性估计

NO.PZ2022122601000045

问题如下:

Board member Arnold Brown asks O'Reilly about the use of high-frequency (daily) data in developing capital market expectations. O'Reilly answers, "Sometimes it is necessary to use daily data to obtain a data series of the desired length. High-frequency data are more sensitive to asynchronism across variables and, as a result, tend to produce higher correlation estimates."

With respect to his answer to Brown's question, O'Reilly most likely is:

选项:

A.incorrect, because high-frequency data are less sensitive to asynchronism B.incorrect, because high-frequency data tend to produce lower correlation estimates C.correct

解释:

Correct Answer: B

O'Reilly's answer is incorrect with respect to correlation estimates. High-frequency data are more sensitive to asynchronism across variables and, as a result, tend to produce lower correlation estimates.

中文解析:

就相关性估计而言,O'Reilly的答案是不正确的。高频数据对变量间的异步更敏感,因此往往产生较低的相关性估计。

老师帮忙解释一下吧,不用太复杂,便于理解记忆就好,谢谢

1 个答案
已采纳答案

源_品职助教 · 2024年02月05日

嗨,爱思考的PZer你好:


高频数据会产生异步性,从而导致低估相关性。

打个比方,比如股票A和B再过去一周都上涨了,相关性表现的很好。

但是我现在把观测周期从一周的时间改为1小时,虽然A和B在过去一周上涨,但是精确到每一个小时,两者的涨跌可能就不同步了。

所以高频的异步性就降低了相关性。

不客气的~

----------------------------------------------
加油吧,让我们一起遇见更好的自己!

  • 1

    回答
  • 3

    关注
  • 330

    浏览
相关问题

NO.PZ2022122601000045问题如下 Boarmember ArnolBrown asks O'Reilly about the use ofhigh-frequen(ily) ta in veloping capitmarket expectations. O'Reillyanswers, \"Sometimes it is necessary to use ily ta to obtain a ta seriesof the sirelength. High-frequenta are more sensitive to asynchronismacross variables an a result, tento prohigher correlation estimates.\"With respetohis answer to Brown's question, O'Reilly most likely is: A.incorrect, because high-frequenta are lesssensitive to asynchronismB.incorrect, because high-frequenta tento procelower correlation estimatesC.correct CorreAnswer: BO'Reilly's answeris incorrewith respeto correlation estimates. High-frequenta aremore sensitive to asynchronism across variables an a result, tentoprolower correlation estimates. 中文解析就相关性估计而言,O'Reilly的答案是不正确的。高频数据对变量间的异步更敏感,因此往往产生较低的相关性估计。 我是把高频交易看作appraista的反面来理解了。由于appraista 会unrestimate correlation 所以高频交易会增加correlation,这样理解有什么问题吗?是因为这道题提到了异步的问题吗?那么在异步的前提下,appraista 是否还是会unrstate correlation呢?

2024-06-30 22:07 1 · 回答

NO.PZ2022122601000045问题如下 Boarmember ArnolBrown asks O'Reilly about the use ofhigh-frequen(ily) ta in veloping capitmarket expectations. O'Reillyanswers, \"Sometimes it is necessary to use ily ta to obtain a ta seriesof the sirelength. High-frequenta are more sensitive to asynchronismacross variables an a result, tento prohigher correlation estimates.\"With respetohis answer to Brown's question, O'Reilly most likely is: A.incorrect, because high-frequenta are lesssensitive to asynchronismB.incorrect, because high-frequenta tento procelower correlation estimatesC.correct CorreAnswer: BO'Reilly's answeris incorrewith respeto correlation estimates. High-frequenta aremore sensitive to asynchronism across variables an a result, tentoprolower correlation estimates. 中文解析就相关性估计而言,O'Reilly的答案是不正确的。高频数据对变量间的异步更敏感,因此往往产生较低的相关性估计。 老师,我知道异步性会使得correlation降低,但是我还有个疑问,就是原本数据少的时候就容易导致correlation被低估,而high-frequency书上又说可以\"improve the precision of sample variances, covarianancorrelations\", 那如果还是低估相关性的话,高频数据对于相关性估计的改进到底改进在哪里呢?

2024-04-19 16:46 1 · 回答