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。

Show HN: A more intuitive linear regression interface (Statwing)

4 点作者 glaugh超过 10 年前

2 条评论

glaugh超过 10 年前
We&#x27;ve been working hard on a more intuitive regression interface:<p>. Plain-English description of results<p>. Inline data transformations (e.g., taking a log)<p>. Automatic diagnostic charts and distribution visualizations<p>. Automatic alerts to common issues (e.g., funky residual plot)<p>. Plain-english guides to regression [1] and interpreting residuals [2]<p>. M-estimation instead of OLS<p>This is of course a work in progress, so we&#x27;d love to get feedback on this. Particularly if you see something in the docs you don&#x27;t agree with.<p>[1] <a href="http://docs.statwing.com/user-friendly-guide-to-regression/" rel="nofollow">http:&#x2F;&#x2F;docs.statwing.com&#x2F;user-friendly-guide-to-regression&#x2F;</a><p>[2] <a href="http://docs.statwing.com/interpreting-residual-plots-to-improve-your-regression" rel="nofollow">http:&#x2F;&#x2F;docs.statwing.com&#x2F;interpreting-residual-plots-to-impr...</a><p>(Logistic regression is also in the works, we&#x27;ll probably do a &quot;Show HN&quot; for that later, so please hold comments on it)
minimaxir超过 10 年前
Interesting choice by linking to a deliberately flawed regression model as an example, which allows the utility of the diagnostics to show.