---
title: "#statistics articles | Hashnode"
description: "Explore 876 articles tagged #statistics on Hashnode. Tutorials, guides, and how-tos from the community."
url: "https://hashnode.com/tag/statistics"
---

Tag

# #statistics

876 posts · 215 followers

[Dustin Liang](https://hashnode.com/@dustinliang) · [1d ago](https://hashnode.com/posts/can-you-trust-a-hurst-exponent-in-python-on-news-attention-data-a-post-mortem-of-r-s-vs-dfa/6afd641e21721d48c56c0f7f)

### [Can You Trust a Hurst Exponent in Python on News-Attention Data? A Post-Mortem of R/S vs DFA](https://dustinliang.hashnode.dev/can-you-trust-a-hurst-exponent-in-python-on-news-attention-data-a-post-mortem-of-r-s-vs-dfa?utm_source=hashnode&utm_medium=feed)

Only partly: a Hurst exponent computed in Python separates persistent from mean-reverting attention series only if you use DFA on log counts, have well over 250 observations, and first check for news

12 min read · [dustinliang.hashnode.dev](https://dustinliang.hashnode.dev/can-you-trust-a-hurst-exponent-in-python-on-news-attention-data-a-post-mortem-of-r-s-vs-dfa?utm_source=hashnode&utm_medium=feed)

[HK Analytical](https://hashnode.com/@hkanalytical) · [1d ago](https://hashnode.com/posts/what-and-k-2-mean-explained-for-engineers/6afd3f5fd8a157f6bb91679d)

### [What ± and k = 2 mean, explained for engineers](https://hkanalytical.hashnode.dev/what-and-k-2-mean-explained-for-engineers?utm_source=hashnode&utm_medium=feed)

A lab report says 9.6 ± 0.4 mg (k = 2). If you work with error bars, latency percentiles or SLOs, you already have most of the mental model. You also have some habits that give wrong answers here. The

6 min read · [hkanalytical.hashnode.dev](https://hkanalytical.hashnode.dev/what-and-k-2-mean-explained-for-engineers?utm_source=hashnode&utm_medium=feed)

[Dustin Liang](https://hashnode.com/@dustinliang) · [2d ago](https://hashnode.com/posts/kalman-filter-in-python-with-statsmodels-smoothing-a-noisy-daily-sentiment-score-without-lagging-real-shifts/6afbbc2c09209622244706d5)

### [Kalman Filter in Python with statsmodels: Smoothing a Noisy Daily Sentiment Score Without Lagging Real Shifts](https://dustinliang.hashnode.dev/kalman-filter-in-python-with-statsmodels-smoothing-a-noisy-daily-sentiment-score-without-lagging-real-shifts?utm_source=hashnode&utm_medium=feed)

A daily sentiment score built from news headlines is noisy for a simple reason: it is an average of very few things. On a busy day you might score ten headlines about a company; on a quiet day, one or

14 min read · [dustinliang.hashnode.dev](https://dustinliang.hashnode.dev/kalman-filter-in-python-with-statsmodels-smoothing-a-noisy-daily-sentiment-score-without-lagging-real-shifts?utm_source=hashnode&utm_medium=feed)

[Sihwan Oh](https://hashnode.com/@neulketing) · [2d ago](https://hashnode.com/posts/how-wide-is-a-3-6-error-bars-for-a-views-vs-median-multiple/6afbb03efa99af78019e80d8)

### [How Wide Is a ×3.6? Error Bars for a Views-vs-Median Multiple](https://webtool-notes.hashnode.dev/how-wide-is-a-3-6-error-bars-for-a-views-vs-median-multiple?utm_source=hashnode&utm_medium=feed)

As of November 17, 2026 Disclosure: I'm on the team that builds TRYIT (free). A Short with 1M views can be a flop if the channel usually gets 3M. So instead of the raw count, we look at a multiple: mu

8 min read · [webtool-notes.hashnode.dev](https://webtool-notes.hashnode.dev/how-wide-is-a-3-6-error-bars-for-a-views-vs-median-multiple?utm_source=hashnode&utm_medium=feed)

[Dustin Liang](https://hashnode.com/@dustinliang) · [2d ago](https://hashnode.com/posts/mann-kendall-trend-test-in-python-is-a-sector-heat-series-really-trending-or-just-autocorrelated/6afba16b3f4854eb07eb35b0)

### [Mann-Kendall Trend Test in Python: Is a Sector Heat Series Really Trending, or Just Autocorrelated?](https://dustinliang.hashnode.dev/mann-kendall-trend-test-in-python-is-a-sector-heat-series-really-trending-or-just-autocorrelated?utm_source=hashnode&utm_medium=feed)

A sector "heat" series (a daily score built from news mentions, search interest or discussion volume) goes up for three weeks and someone asks the obvious question: is this a trend, or is it just nois

13 min read · [dustinliang.hashnode.dev](https://dustinliang.hashnode.dev/mann-kendall-trend-test-in-python-is-a-sector-heat-series-really-trending-or-just-autocorrelated?utm_source=hashnode&utm_medium=feed)

[Prachit Parikh](https://hashnode.com/@prachitparikh) · [2d ago](https://hashnode.com/posts/ai-uat-sample-size/6afb67bd3c2fd5c18b1fe613)

### [AI UAT sample size: zero failures in 20 runs isn't a pass, so here's how many runs you need](https://prachitparikh.hashnode.dev/ai-uat-sample-size?utm_source=hashnode&utm_medium=feed)

Zero failures in 20 runs still allows a 14% failure rate. How to size UAT for an AI feature, with a 30-line Python script and a sign-off table.

19 min read · [prachitparikh.hashnode.dev](https://prachitparikh.hashnode.dev/ai-uat-sample-size?utm_source=hashnode&utm_medium=feed)

[dragonair](https://hashnode.com/@dragonair) · [4d ago](https://hashnode.com/posts/spc-charts-people-actually-keep-using/6af9286bb9dc05947b08d481)

### [SPC charts people actually keep using](https://6sqtools.hashnode.dev/spc-charts-people-actually-keep-using?utm_source=hashnode&utm_medium=feed)

Most control charts die within a quarter. The usual cause is not the operator - it is a chart that was set up to be administratively convenient instead of statistically valid.

4 min read · [6sqtools.hashnode.dev](https://6sqtools.hashnode.dev/spc-charts-people-actually-keep-using?utm_source=hashnode&utm_medium=feed)

[dragonair](https://hashnode.com/@dragonair) · [4d ago](https://hashnode.com/posts/your-cpk-looks-great-and-the-process-still-fails/6af925f407dd15f978924670)

### [Your Cpk looks great and the process still fails](https://6sqtools.hashnode.dev/your-cpk-looks-great-and-the-process-still-fails?utm_source=hashnode&utm_medium=feed)

Cpk and Ppk answer different questions. Mixing up the two sigmas is the most common reason a capability report turns out meaningless.

4 min read · [6sqtools.hashnode.dev](https://6sqtools.hashnode.dev/your-cpk-looks-great-and-the-process-still-fails?utm_source=hashnode&utm_medium=feed)

[Neu Portal](https://hashnode.com/@neuportal) · [5d ago](https://hashnode.com/posts/how-to-tell-whether-your-forecasting-model-beats-the-market-a-python-walkthrough/6af786fdc9c4a9fa707c8f6c)

### [How to Tell Whether Your Forecasting Model Beats the Market: A Python Walkthrough](https://neuportal.hashnode.dev/how-to-tell-whether-your-forecasting-model-beats-the-market-a-python-walkthrough?utm_source=hashnode&utm_medium=feed)

Brier scores, the Murphy decomposition, a clustered bootstrap and a blending test, run on an open dataset of 5,654 timestamped model forecasts against Polymarket prices. If you build a probabilistic m

7 min read · [neuportal.hashnode.dev](https://neuportal.hashnode.dev/how-to-tell-whether-your-forecasting-model-beats-the-market-a-python-walkthrough?utm_source=hashnode&utm_medium=feed)

[Mohammed M.sharaki](https://hashnode.com/@Mo7ammedSharaki) · [5d ago](https://hashnode.com/posts/week-i-of-learning-ml-as-statistics-student/6af772dee6dc2e62fd359fd6)

### [Week-I of learning ML as STATISTICS student.](https://mosharaqi.hashnode.dev/week-i-of-learning-ml-as-statistics-student?utm_source=hashnode&utm_medium=feed)

Welcome everyone to \\(WEEK\_n\\), where I share my journey of learning ML from scratch and here i'm sharing my journey with you. I can confidently say that I had a great introduction to AI last summer.

2 min read · [mosharaqi.hashnode.dev](https://mosharaqi.hashnode.dev/week-i-of-learning-ml-as-statistics-student?utm_source=hashnode&utm_medium=feed)
