Foundations in Statistics Ep. 5 | Measurement Error скачать в хорошем качестве

Foundations in Statistics Ep. 5 | Measurement Error

This episode is a deep dive into the hidden reality of the physical world: The fact that no measurement is ever truly precise. We explore why "all measurements are wrong" and how we use statistics to handle this uncertainty anyway. In This Episode... Master Equation of any Measurement: Individual Measurement = True Value + Bias + Chance Error We look at the fascinating case of NB 10, a 10-gram checkweight measured by NIST for over 80 years, to see how random errors cluster and why the Standard Deviation (SD) is the most powerful tool for estimating the size of your error. Key Topics: The Myth of Exactness: Why physical traits are "continuous" and our instruments are "discrete." Chance Error vs. Bias: Why you can't fix a "heavy thumb" by taking more measurements The Ethics of Outliers: When is it okay to throw out a "bad" data point? (The NIST approach) Precision vs. Accuracy The Quest for the Kilogram: The story of "Le Grand K" in Paris and the 2019 shift to Planck’s Constant Brief intro to Gage R&R Studies: How industries determine if a measurement system is "good enough" for the job Main Reference: Our primary source for this series is "Statistics" by David Freedman et al. This book is one of the best resources for understanding the logic of statistics with minimal math and maximum intuition. Timestamps: 0:00 The Jeweler's Dilemma 1:14 Why Measurement Error Matters 2:15 "All Measurements are Wrong" (The Concept of Continuous Variables) 3:30 The Physical World is Inherently Random 4:50 The Language of Uncertainty 6:25 The Master Equation: Truth, Bias, and Chance Error 8:30 What is the Chance Error? (The Microscopic Forces) 11:00 How Big is the Chance Error? Case Study: NB10 & The National Bureau of Standards 16:30 Standard Deviation: Your Chance Error Compass 20:10 Outliers in NB10 Case Study 21:30 The Outlier Problem 23:50 Understanding Bias in Measurement 26:20 Precision vs. Accuracy (The Tight Cluster vs. The Bullseye) 27:45 Fixing Bias: Calibration Chain: Traceability, History of Le Grand K & The 2019 Redefinition 31:30 Introduction to Gage R&R Studies 32:50 Summary #Statistics #MeasurementError #Engineering #DataScience #StandardDeviation #FreedmanStatistics #Metrology #QualityControl #NIST #SixSigma #MathSimplified