Calculus 1000-Second Derivative Test-(Section 007, November 11 Fall 2026)
Overview
Asghar Ghorbanpour develops the second derivative test by connecting the sign of f'' to concavity, then uses sign charts to find concavity intervals and points of inflection. He states how f'' at a critical point identifies a local maximum or minimum, explains why f'' = 0 is inconclusive, and compares the test’s pointwise calculation with the first derivative test’s interval sign chart.
Key takeaways
- For a twice-differentiable function, f'' > 0 on an interval means concave up, while f'' < 0 means concave down; a sign chart for f'' locates the concavity intervals.
- A zero of f'' is only a candidate for an inflection point: concavity must actually change across the point.
- A continuous function can have an inflection point where f' is undefined, as ∛x does at x = 0.
- At a critical point c with f'(c) = 0, f''(c) < 0 guarantees a local maximum and f''(c) > 0 guarantees a local minimum.
- When f''(c) = 0, the second derivative test is inconclusive; x³, x⁴, and −x⁴ demonstrate that the point may be neither extremum, a minimum, or a maximum.
- The first derivative test needs signs of f' on intervals around a critical point, whereas the second derivative test can classify it by evaluating f'' at that single point when the value is nonzero.
Chapters
- The first derivative test identifies a local maximum from an increasing-then-decreasing pattern and a local minimum from a decreasing-then-increasing pattern.
- Ghorbanpour introduces the second derivative test as another way to classify local extrema and previews concavity as the link.
- On an interval, a differentiable function is concave upward when f' is increasing and concave downward when f' is decreasing.
- The graph’s curvature direction provides a visual guide: upward curvature corresponds to increasing slopes, while downward curvature corresponds to decreasing slopes.
- For a twice-differentiable function, f'' > 0 throughout an interval means f' is increasing, so the function is concave up there.
- If f'' < 0 throughout an interval, f' is decreasing and the function is concave down.
- Finding concavity intervals therefore reduces to making a sign chart for f''.
- For f(x) = x⁴ − 4x³ − 18x² + 12x − 3, differentiation gives f'(x) = 4x³ − 12x² − 36x + 12 and f''(x) = 12x² − 24x − 36.
- Factoring f'' gives 12(x − 3)(x + 1), so the sign-chart boundaries are x = −1 and x = 3.
- The sign pattern is positive, negative, positive, giving concave-up intervals (−∞, −1) and (3, ∞), and a concave-down interval (−1, 3).
- For g(x) = arctan(x), g'(x) = 1/(1 + x²) and g''(x) = −2x/(1 + x²)².
- The denominator of g'' is always positive, so its sign is determined by −2x: positive for x < 0 and negative for x > 0.
- Thus arctan(x) is concave up on (−∞, 0) and concave down on (0, ∞), matching the curvature change in its graph.
- For h(x) = x/(x² + 1), the second derivative has zeros at x = −√3, 0, and √3.
- The sign of h'' alternates across those three values, producing concave-down intervals (−∞, −√3) and (0, √3), and concave-up intervals (−√3, 0) and (√3, ∞).
- The corresponding function values at the nonzero boundaries are h(−√3) = −√3/4 and h(√3) = √3/4; h(0) = 0.
- A point of inflection occurs where the function’s concavity changes; f'' = 0 alone does not establish an inflection point.
- For the quartic, the sign changes at x = −1 and x = 3, so both are inflection-point x-coordinates; for arctan(x), the change occurs at x = 0.
- For h(x) = x/(x² + 1), concavity changes at x = −√3, 0, and √3, giving inflection points (−√3, −√3/4), (0, 0), and (√3, √3/4).
- The broader criterion discussed is that the function must be continuous at the candidate point and its concavity must change there.
- A point of inflection does not require the first derivative to exist: the cube-root function ∛x has an inflection at x = 0 even though its derivative is undefined there.
- A candidate where the function itself is discontinuous cannot be called a point of inflection under this definition.
- At a critical point c where f'(c) = 0, f''(c) < 0 implies a local maximum and f''(c) > 0 implies a local minimum.
- If f''(c) = 0, the test is inconclusive: examples such as x³, x⁴, and −x⁴ have different behaviors at x = 0 despite the zero second derivative.
- The test classifies candidates; identifying the critical point comes first.
- The first derivative test requires a sign chart for f' on intervals to either side of a critical point and can classify a point as a maximum, minimum, or neither.
- The second derivative test only requires evaluating f'' at the critical point, which can avoid building an interval sign chart when f'' is nonzero there.
- For the example with f'(x) = (eˣ − 1)/((eˣ − x)² + 1), solving f'(x) = 0 gives the critical point x = 0; the proposed second-derivative approach evaluates f'' only at that point.
- Ghorbanpour emphasizes the main tradeoff: the second derivative test can be quicker at a critical point, but f''(c) = 0 provides no classification.
- The lecture closes the discussion of derivative-based maximum and minimum tests and previews optimization examples for the next class.
Transcript excerpt
0:00 Okay, hello everyone. Welcome back to the calculus. Today, we will continue with what we started. So we are trying to see how derivatives and we want to understand how derivatives show some information about the graph of the function. But along the way, there is also a parallel goal that comes with us, which is kind of going towards finding maximum and minimum. And but so far we have been able to settle down the local max using the first derivative test.
0:50 Well, the key idea there was that we noticed that if something a point is going to give us local max, the function will have increasing, decreasing pattern before after that point. And if it is going to be local min, it's going to be first decreasing, then increasing. So that was the key point, very simple observation, and worked out well for us. But there is also another point that we we can make there, and this is now we want to use the second derivative to kind of identify the local max min. So this is called second derivative test.
1:39 Well, we will start with a little bit of of what is the meaning of the second derivative for the graph, and then we go move on for the the second group. So that's the plan for today. Any questions, comments before we start? So let's start with the definition of what does it mean a function being concave up or down? So this is
2:26 concave up versus down. Well, concavity of a function can be defined in different ways. [snorts] Like yesterday the discussion Oh, somebody erased it. The discussion we had actually gave us two good ways of identifying that. There is actually one definition that doesn't need the function to be differentiable. It is a
This is the start of the transcript. Save this video free for the whole transcript, searchable and time-coded.
In this course
Related notes
Summary, takeaways, and chapters were generated by AI from the video's transcript and may contain errors. The video belongs to its creator, Asghar Ghorbanpour.