TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Poisson designs and minimum detectable effects

13 点作者 apwheele10 个月前

1 comment

cl3misch10 个月前
&gt; Counter-intuitive perhaps, you get smaller standard errors when you observe higher counts.<p>No, the <i>relative</i> error aka. std&#x2F;mean decreases. For Poisson, var=mean=N =&gt; std&#x2F;mean = sqrt(N)&#x2F;N = 1&#x2F;sqrt(N). I&#x27;m sure the author knows this but the post is not written very clearly imho. I wonder who the target audience is?<p>When I was a junior researcher I rather found confusing that variance <i>increases</i> with increased counts. But the perceived &quot;error&quot; is the standard deviation relative to mean, which does decrease as shown above.
评论 #41103721 未加载
评论 #41104025 未加载