Calculus Pop Quiz

Calculus Level 3

See how well you understand differentiability!

Consider a differentiable function f : R R f : \mathbb{R} \rightarrow \mathbb{R} with f ( 0 ) = 0 f(0)=0 and f ( 0 ) = 1 f'(0)=1 . How many of the following statements are true for any such function f f ?

1) f ( x ) > 0 f(x)>0 on ( 0 , q ) (0,q) for some positive q q ,

2) f ( x ) f(x) is increasing on ( p , q ) (p,q) for some negative p p and some positive q q

3) f ( x ) |f(x)| is continuous,

4) There exists a differentiable function g : R R g: \mathbb{R} \rightarrow \mathbb{R} such that g ( x ) = f ( x ) g''(x)=f(x) , and

5) f ( x ) f'(x) is continuous.


I recently gave this pop quiz in week three of an introductory calculus class; it did not go so well. Let's see whether the "Brilliant class" can do better ;)


The answer is 3.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

3 solutions

The correct statements are 1, 3, 4.


Subparagraph before the solution:

Differentiable function is not the same as derivative function, at all. If f : R n R f:\mathbb{R}^n \to \mathbb{R} is a differentiable function then it has directionable derivates ,nevertheless a function is able to have a derivate on every direction and it has not to be a differentiable function. Over the compex numbers also are different concepts,...

However, in this case f : R R f:\mathbb{R} \rightarrow \mathbb{R} is a differentiable function if and only if it is a derivative function.

Proof

\Rightarrow ) Let's suppose that f : R R f:\mathbb{R} \to \mathbb{R} is a differentiable function, then c R \forall c \in \mathbb{R} f ( x ) f ( c ) = d f c ( x c ) + x c ρ ( x c ) f(x) - f(c) = df_{c} (x - c) + |x - c|\rho(x - c) x \forall x in a reduced neighborhood of c(neighborhood without c), where d f c df_{c} is a R \mathbb{R} -linear transformation and ρ ( h ) \rho(h) is a function defined in a reduced neighborhood of c fulfilling ρ ( h ) 0 \rho(h) \to 0 as h 0 h \to 0 which implies that f ( x ) f ( c ) = ( x c ) d f c ( 1 ) + x c ρ ( x c ) lim x c f ( x ) f ( c ) x c = d f c ( 1 ) . . . f(x) - f(c) = (x - c)df_{c} (1) + |x - c|\rho(x - c) \Rightarrow \lim_{x \to c} \frac{f(x) - f(c)}{x - c} = df_{c}(1)...

\Leftarrow ) Let's suppose that f : R R f:\mathbb{R} \to \mathbb{R} is a derivative function, then c R \forall c \in \mathbb{R} lim x c f ( x ) f ( c ) x c = f ( c ) \lim_{x \to c} \frac{ f(x) - f(c)}{x - c} = f '(c) \Rightarrow lim x c f ( x ) f ( c ) x c f ( c ) = 0 f ( x ) f ( c ) f ( c ) ( x c ) = o ( x c ) \lim_{x \to c} \frac{ f(x) - f(c)}{x - c} - f '(c) =0 \Rightarrow f(x) - f(c) - f '(c)(x - c) = o(|x - c|) \Rightarrow f ( x ) = f ( c ) + f ( c ) ( x c ) + o ( x c ) f(x) = f(c) + f '(c)(x - c) + o (|x - c|) in a reduced neighborhood of c.... Read note below \square .


Solution

Consider a differentiable function f : R R f:\mathbb{R} \to \mathbb{R} such that f ( 0 ) = 0 f(0) = 0 and f ( 0 ) = 1 f '(0) = 1 .

1 \boxed{1} .- f ( x ) = f ( 0 ) + f ( 0 ) x + o ( x ) f(x) = f(0) + f '(0)x + o(|x|) in a neighborhood of 0, which implies that q > 0 / f ( x ) 0 + 1 x x x ( 0 , q ) f ( x ) > 0 x ( 0 , q ) \exists q > 0 \space / f(x) \approx 0 +1 \cdot x \approx x \space \forall x \in (0,q) \Rightarrow f(x) > 0 \space \forall x \in (0,q) .

2 \boxed{2} .- The stament 2 and 5 are false . Let f ( x ) = x 2 x 2 sin ( 1 x ) , if x is distinct to 0 and, f ( 0 ) = 0 f(x) = x - 2x^2\sin(\frac{1}{x}) \text{, if x is distinct to 0 and, } f(0) = 0 . This function fulfills f ( 0 ) = 0 , f ( 0 ) = 1 f(0) = 0, f '(0) = 1 is differentiable over R \mathbb{R} with not continuous derivate(it is not continuous on 0) and not fulfilling the statement 2 because in all neighborhood of 0 there is a point x where f ( x ) < 0 f '(x) < 0 \Rightarrow the function is not increasing in any neighborhood of 0.

This example is also interesting because if f : R n R f:\mathbb{R}^n \to \mathbb{R} has partial derivates functions being continuous then f is a differentiable function. And this example shows that the reciprocal is not true.

3 \boxed{3} .- The absolute value function is a continuous function and the composition of two continuous functions is a continuous function. Every differentiable function f : R R f:\mathbb{R} \to \mathbb{R} is a continuous function: a R , lim x a f ( x ) = lim x a f ( a ) + f ( a ) ( x a ) + o ( x a ) = f ( a ) \forall a \in \mathbb{R}, \lim_{x \to a} f(x) = \lim_{x \to a} f(a) + f ' (a)(x -a) + o (|x -a|) = f(a) \Rightarrow ϵ > 0 , δ > 0 / f ( x ) f ( a ) < ϵ if 0 x a < δ \space \forall \epsilon > 0, \exists \delta > 0 \space / \space |f(x) - f(a)| < \epsilon \text{ if } 0 \leq |x - a| < \delta

4 \boxed{4} .- All continuous function is Riemann integrable, and due to Fundamental Calculus Theorem h ( x ) = 0 x f ( t ) d t h(x) = \int_{0}^{x} f(t) dt fulfills h ( x ) = f ( x ) h ' (x) =f(x) . Take g ( x ) = 0 x h ( t ) d t g ( x ) = h ( x ) g ( x ) = h ( x ) = f ( x ) g(x)=\int_{0}^{x} h(t) dt \Rightarrow g '(x) = h(x) \Rightarrow g ''(x) = h '(x) = f(x)

Note .- o ( x ) = x o ( 1 ) = x ρ ( x ) o(|x|) = |x|o(1) = |x|\rho(x) where ρ ( x ) \rho(x) is a function with ρ ( x ) 0 \rho(x) \to 0 as x 0 x \to 0 ( o ( x ) o(|x|) means lim x 0 o ( x ) x = 0 \lim_{x \to 0} \frac{o(|x|)}{x} = 0 )

Wow, I'm delighted to find all this wonderful work here as I wake up in the morning! Thank you Pranav and Guillermo!

The trickier parts are (2) and (5), which are related: There exist differentiable functions whose derivative fails to be continuous. Guillermo gives a great example. We should mention that his function is piecewise: f ( x ) = x 2 x 2 sin ( 1 x f(x)=x-2x^2\sin(\frac{1}{x} ) for nonzero x x and f ( 0 ) = 0 f(0)=0 . Note that f ( 1 2 k π ) = 1 f'(\frac{1}{2k\pi})=-1 for odd k k .

For those who are uncomfortable with the "little o" notation, here is another approach to part (1): We are told that f ( 0 ) = lim h 0 f ( h ) f ( 0 ) h = lim h 0 f ( h ) h = 1 f'(0)=\lim_{h\to 0}\frac{f(h)-f(0)}{h}=\lim_{h\to 0}\frac{f(h)}{h}=1 . Using the definition of a limit with ϵ = 1 \epsilon=1 , we conclude that there exists a δ > 0 \delta>0 such that f ( h ) h > 0 \frac{f(h)}{h}>0 for 0 < h < δ 0<|h|<\delta , which implies that f ( h ) > 0 f(h)>0 on the interval ( 0 , δ ) (0,\delta) .

Otto Bretscher - 5 years, 2 months ago

Log in to reply

Thanks for you Otto, you taught me the statements 2 and 5 are false, I really appreciate this, thank you, my pelasure...

Guillermo Templado - 5 years, 2 months ago

Log in to reply

The pleasure is mine ;)

Otto Bretscher - 5 years, 2 months ago

Log in to reply

@Otto Bretscher @Guillermo Templado @Otto Bretscher Fixed.

I also copied the subsequent conversation over.

If you subscribe to comments, you will be notified about new posts.

Calvin Lin Staff - 5 years, 2 months ago

Log in to reply

@Calvin Lin Thank you so much @Calvin Lin for fixing this!

Otto Bretscher - 5 years, 2 months ago

@Calvin Lin Thank you Calvin, What about a wiki about "Lagrange multipliers and minimum squares methodus"? PS.- As usual, I write it and you later correct my edits... Ok, I'm going soon to start this wiki, but this wiki is going to be very long, there will be proofs and examples, and it'll take me very long, I'll email you about my progress for your review

Guillermo Templado - 5 years, 2 months ago

Log in to reply

@Guillermo Templado That's great! Let me know if you need help with brainstorming what the sections would be like.

lagrange multipliers is definitely a concept that many people are interested in, but only get a faint grasp of it initially. Most forget to check the boundary conditions, or the second order conditions, because they don't understand the ideas/theory behind it.

Calvin Lin Staff - 5 years, 2 months ago

Log in to reply

@Calvin Lin I'll continue this wiki, Lagrange multipliers ,with some examples first, and we'll talk with e-mails about the progress.I hope not to bother to Samarth... My demonstration about Lagrange multipliers theorem really is not easy, it needs implicit function theorem and a lot machinery and is very long and is not accesible for everybody... Ok, I'll contact you and we`ll talk...

Guillermo Templado - 5 years, 2 months ago
Otto Bretscher
Mar 22, 2016

Request for @Calvin Lin : Our Compañero @Guillermo Templado posted a wonderful solution as a comment, but unfortunately that solution was deleted, along with the comments. Is there any way to bring Guillermo's comment back as a solution, with his consent? It was a long post, and it would be a shame for all that careful work to be lost.

I'm extremely sorry. When I saw the solution by Guillermo Templado, I understood my mistakes. So without thinking twice ,I just deleted my comment before realising that even the comments by others will be deleted.It was really a long solution. I'm really sorry. And a great sum. Thanks for posting it.

Pranav Rao - 5 years, 2 months ago

Log in to reply

Don't worry, Pranav. It will be fixed... be calm

Guillermo Templado - 5 years, 2 months ago

Indeed, no worries... it's all good now!

Otto Bretscher - 5 years, 2 months ago

considering the function to be y=(sinx) then we will not get the first statement true.

Why not? Let q = π q=\pi

Otto Bretscher - 5 years, 1 month ago

Log in to reply

but if we put q=2pi then 1st statement would be incorrect.

Siddharth Tripathi - 5 years, 1 month ago

Log in to reply

It says "for some positive q q ", not for all.

Otto Bretscher - 5 years, 1 month ago

Log in to reply

@Otto Bretscher thanks for clearing the doubt

Siddharth Tripathi - 5 years, 1 month ago

Log in to reply

@Siddharth Tripathi my pleasure

Otto Bretscher - 5 years, 1 month ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...