Library / Linear / Linear Regression

Linear Regression

Best-fit line plus a causation audit. A big coefficient is not a big cause.

outlier pulls line
best-fit line ≠ cause · recreated schematic
INTERACTIVE LAB

Tolerance sandbox (original recreation)

happy 76% • avg. similar neighbors 49% • red ring = wants to move

LINEAR

When to use / misuse

Use → Screen relationships, then hunt omitted variables and reverse causality.

Difficulty → Starter

REDCAPE → Reason · Predict

agent: linear-regression-checker
family: Linear
combine_with: [two other families]
rule: state failure mode before acting

Free sources: Coursera M3 (Categorical & Linear) • OpenIntro Statistics

← LibrarySee Playbook →
◍ Ask mentor about this →See REDCAPE playbook →

Attach to a lattice — combines with

Different families, shared jobs. Click to open alongside.

DISTRIBUTIONS
Normal Distribution
Bell-curve averaging: many small independent effects sum to a predictable center with thin tails.
SHARED JOBS: REASON · PREDICT
DYNAMICS
Random Walk & Markets
Unpredictable steps accumulate; efficient markets leave mostly randomness behind.
SHARED JOBS: REASON · PREDICT
NETWORKS
Schelling Segregation
Mild individual preference for similar neighbors aggregates to stark city-wide segregation.
SHARED JOBS: REASON · PREDICT

More in Linear

Concavity & Diminishing Returns
Curved payoffs change optimal bets: diversify when concave, concentrate when convex.
Categorical & Threshold Models
Buckets and cutoffs turn messy attributes into testable if-then rules.
← Poisson ArrivalsConcavity & Diminishing Returns →