Univariate Chain Rule
The familiar single-variable chain rule says: if , then .
In terms of Jacobians: . Since , all Jacobians are matrices (scalars), and Jacobian multiplication is just scalar multiplication. The univariate chain rule is the matrix case of the general chain rule.
The key insight from this perspective: the intermediate variable appears as the evaluation point for . This is why we write , not — we must differentiate at the intermediate value , not at itself.
Formal View
In Leibniz notation: where . The "cancellation of " is a useful mnemonic, though not literally true.
Interactive Visualization
Chain Rule Circuit Diagram
Why This Matters
The univariate chain rule is the foundation for all differentiation of compositions — mastering it is essential.
- Differentiating , , and all composed scalar functions
- Implicit differentiation: treating as a function of in
- Solving ODEs by substitution: transforms the equation
Quiz
If , then equals:
In the chain rule , where is evaluated?
Common Mistakes
- Evaluating at instead of at — the outer derivative must be evaluated at the intermediate variable.
- Forgetting to multiply by — every chain rule application adds a factor.
- Confusing the chain rule with the product rule for .