Linear models and simulations | MIT Computational Thinking Spring 2021 | Lecture 15

preview_player
Показать описание

Contents
00:00 Introduction
01:30 Fahrenheit and Celsius Data Set
02:45 Julia: Data Frames
03:10 Data Frame by Columns with labels
04:23 Data Frame with a matrix
05:25 Reading/Writing CSV (comma separated values) Files
08:00 Noisy Data: Add some random noise to the celsius readings
12:15 Statistics Software Outputs Mysterious Tables
14:29 Regression a few ways
15:53 The "Coef." column in the table gives the slope and intercept of the best fit line
21:33 Demystifying the word "Model"
25:20 Understanding the relationship °C ~ 1 + °F
27:05 Simulating the real world: running many noisy models
27:58 Julia: underscore as a digits separator
28:15 Simulated intercepts (100000 simulations)
33:05 Simulated slopes (100000 simulations)
34:55 Simulated σ (100000 simulations)
40:15 The Linear Model Table
41:28 The Coef column is just the regression formula for the best line
41:38 The Std. error column
44:05 The t column
45:30 The t-distribution
47:45 The ??? column
51:00 Degrees of Freedom

Рекомендации по теме
Комментарии
Автор

humm... why Julia professor Ellen not showing up

hsqgttt
Автор

Where can I find the explanation of the formula for the slope estimate ?
mᵉ = sum( x0 .* y0 ) / sum( x0.^2 )

er
Автор

Can we find these notebooks somewhere?

cocoabutter