Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

The Fundamental Theorems of Calculus

  1. Mar 8, 2006 #1
    Can someone break down these theorems for me please because my book is horrible at explaining them. The examples the book gives shows the initial question but then the answer and none of the steps in between even on the very simple questions. I'm confused.

    ~Kitty
     
  2. jcsd
  3. Mar 8, 2006 #2

    Galileo

    User Avatar
    Science Advisor
    Homework Helper

  4. Mar 8, 2006 #3
    I understand what it is but what does it mean? What about the other theorems?

    ~Kitty
     
  5. Mar 8, 2006 #4

    Galileo

    User Avatar
    Science Advisor
    Homework Helper

    It means that differentiation and integration are, in a way, inverse processes. The fundamental theorem only says that in a much more precise way.

    And what other theorems are you talking about? There are 2 parts to the fundamental theorem and both are explained (with an step-by-step proof) on the link.
     
  6. Mar 9, 2006 #5
    My book says there is a second fundamental theorem of calculus.

    ~Kitty
     
  7. Mar 9, 2006 #6
    The second fundamental theorem of calculus holds for f a continuous function on an open interval I and a any point in I, and states that if F is defined by:
    [tex]F(x) = \int_a^xf(t)dt[/tex]
    Then,
    [tex]F'(x) = f(x)[/tex]
    at each point in I, where F'(x) is a derivative of F(x).
     
  8. Mar 9, 2006 #7
    I didn't understand this one either. My book says even less than that. :frown:

    ~Kitty
     
  9. Mar 9, 2006 #8

    mathwonk

    User Avatar
    Science Advisor
    Homework Helper

    the point is that every continuous function is the derivative of some other function. and more precisely that other function can be defined as a limit of riemann sums (:"riemann integral") of the first function.

    intuitively, the moving area function under the graph of f, is an antiderivative of f, if f is continuous.

    the other fundamental theorem is really the mean value theorem, which says that two differentiable functions with the same derivative on an interval, differ by a constant on that interval.

    it follows that for any antiderivative F of f on the interval [a,b], F(x) differs by a constant from the riemann integral of f from a to x. this trivial corollary of the previous two theorems is somewhat ridiculously called the second fundamental theorem of calculus.

    so the main points are how to construct a function witha given derivative, and then how far a function is determined just by knowing its derivative.


    for instance, any riemann integrable function si continuous almost everywhere, hence its riemann integral is an antiderivative alm,ost everywhere.

    but if you are given a function F that has derivative equal to an integrable f almost everywhere, it does not follow that F differs from the riemann integral of f by a constant.

    one needs also to assume that F is lipschitz continuous. i.e. the refined mean value theorem does not say that a function F whose derivative is zero almost everywhere is constant, but does say that if the function F is also lipschitz continuous.

    thus the two parts of the fundamental theoprem concern
    1) the fact that an integral of f is also an antiderivative of f,
    and
    2) the extent to which that property characterizes the integral of f.

    thus as usually stated the 1st theorem says if f is continuous, then the integral is an antiderivative of f everywhere,

    and the 2nd theorem says that conversely, if f is continuous on [a,b] and F is an antiderivative of f on [a,b], then F(x) - F(a) equals the integral of f from a to x.
     
    Last edited: Mar 9, 2006
  10. Mar 9, 2006 #9
    Wow...I need a minute to process that... :bugeye:

    ~Kitty
     
  11. Mar 9, 2006 #10

    mathwonk

    User Avatar
    Science Advisor
    Homework Helper

    only a minute? it took me 40 years to learn that.
     
  12. Mar 12, 2006 #11
    lol, you must be like 500 years old.

    Suppose you unroll a carpet that in not rectangular, but it's edges can form different shapes like a parabola, or anyother function, then as you unroll this carpet the rate of change of the area being swept is exactly its width at any particular instance.
     
  13. Mar 12, 2006 #12

    mathwonk

    User Avatar
    Science Advisor
    Homework Helper

    what you said took me a little less, maybe 10 seconds. its what i said that took longer. maybe you should read it again. but you are right, i am 500 years old.
     
    Last edited: Mar 12, 2006
  14. Mar 13, 2006 #13
    40 years? Do you mean that you lacked an understanding of the Fundamental Theorem of Calculus for 40 years after you first encountered it?

    I don't think this so...but is this what you meant?
     
  15. Mar 14, 2006 #14
    Lol, I think its safe to say that it is what he what he meant, but it was in a joking manner.

    I understand it I think what confuses me is the proof of this theorem.

    ~Kitty
     
  16. Mar 14, 2006 #15

    arildno

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    Dearly Missed

    This is indeed, a crucial theorem, and unfortunately, it seems that many tend to forget it as one of the two pillars upon which applied calculus is built.
     
  17. Mar 14, 2006 #16
    Math, I really don't see how this theorem is really the mean value theorem. Isn't the mean value theorem f(c)(b-a) which is really (1/b-a)(b-a)? Grr (I get furstrated when I get confused)...getting confused......:devil:

    ~Kitty
     
  18. Mar 14, 2006 #17

    mathwonk

    User Avatar
    Science Advisor
    Homework Helper

    well phooey, i have been typing for about 15 minutes and all of a suddeen the browser erased all of my work.


    anyway i am not kidding. if you understabnd everything i have written and think it ridiculous that ti took me 40 to learn it, more power to you.

    but if you think the ftc just says a continuous functiuon is differentiable everywhere and has derivative equal to the original fucntion, and any antiderivative differs from the origiinal continuous function by a constant then you know diddly.
     
  19. Mar 17, 2006 #18

    VietDao29

    User Avatar
    Homework Helper

    D'oh, what a stupis browser... It prevents me from learning something new, and understand the ftc more deeply... :frown:
     
  20. Mar 17, 2006 #19

    mathwonk

    User Avatar
    Science Advisor
    Homework Helper

    i also had a marvellous proof of the riemann hypothesis but that was erased too!!
     
  21. Mar 17, 2006 #20

    mathwonk

    User Avatar
    Science Advisor
    Homework Helper

    tive


    oh no!!! it happened again!!! this si too much. i rewrote the whole integral business explanation, this is frustrating.

    i didn't even get the chance to savem, it ahppoened while i was astill wrioting.

    this never happened i might add before this new skin was created and forced on us. (dig, dig.)
     
  22. Mar 17, 2006 #21

    HallsofIvy

    User Avatar
    Science Advisor

    Basically, the fundamental theorems say that integration and differentiation are (almost) inverses:

    If F(x) is an integral of f(x) then f(x) is the derivative of F'(x):
    [tex]\frac{d}{dx}\int f(x)ds= f(x)[/tex]

    If f(x) is the derivative of F(x) then F(x) is an anti-derivative of f(x):
    [tex]\int \frac{dF}{dx} dx= F(x)+ constant[/tex].

    That constant is the reason for the "almost" above.
     
    Last edited by a moderator: Mar 17, 2006
  23. Mar 17, 2006 #22
    Right. Doesn't adding C just mean that the antiderivative belongs to a family of antiderivatives that are all verticle translations of one another?

    ~Kitty
     
  24. Mar 17, 2006 #23
    By the way Mathwonk, I'm really sorry the browser erased all of your work. Thats horribly crummy. :frown:

    ~Kitty
     
  25. Mar 17, 2006 #24
    mathwonk, I've developed the habit of selecting everything I write in a post (ctrl+A) every few minutes if I'm writing a lengthy post, and copying it (ctrl+C). That way if the browser crashes or I'm forced to relogin to post I can simply paste with minimal losses.

    Maybe you should consider picking up this habit too seeing as your long posts are always excellent. :smile:
     
  26. Mar 17, 2006 #25

    mathwonk

    User Avatar
    Science Advisor
    Homework Helper

    with your kind encouragement I have tried typing the post into a wqord document and pasting:

    third try:

    the two fundamental theorems of calculus are about the relation betwen two ideas: integrals and antiderivatives. Contrary to popular belief they are not the same, but are closely related. or rather for them to be the same in general, one must define antiderivative more carefully than is usually done.

    basically the first FTC says to what extent an integral is also an antiderivative, and the second FTC says to what extent an antiderivative is also an integral.

    For example, for continuous functions f, antiderivatives and integrals are essentially the same, but why restrict the discussion to continuous functions, when many other functions also have integrals? I.e. to what extent is the integral of a non continuous function also an antiderivative?, and to what extent is the antiderivative of a non continuous function also an integral?


    The usual theorems are:

    1) If f is continuous on [a,b] then for every x in [a,b] f is also Riemann integrable on [a,x] and the function F defined by F(x) = Riemann integral of f on [a,x], is an antiderivative of f on [a,b]. I.e. F’(x) = f(x) for all x in [a,b].

    2) If f is continuous on [a,b], and if G is any diferentiable function on [a,b] such that G’(x) = f(x) for all x in [a,b], then G differs from the previously defined F by a constant, i.e. G(x) -G(a) = F(x), for all x in [a,b].

    The proof of the second theorem is obtained by combining the first theorem with the mean value theorem, i.e. given the first theorem, the second one is really the mean value theorem, which tells you the relation between two functions having the same derivative.

    But the theorems are more interesting if f is only assumed Riemann integrable but not continuous. We say a statement about [a,b] is true almost everywhere, if for each e>0, the set where it is not true can be covered by a sequence of intervals of total length less than e.

    1)A If f is Riemann integrable on [a,b], then f is continuous almost everywhere, hence the function F(x) = integral of f on [a,x], is differentiable with derivative equal to f almost everywhere.
    (This is due to Riemann.)

    Suprisingly, it is NOT true that any function G which is differentiable with derivative equal to f almost everywhere on [a,b] differs from F by a constant, not even if we assume G is continuous. Namely there exists a (Cantor) function G continuous on [a,b] and with derivative equal to zero almost everywhere, and yet with G not only not constant, but with G increasing weakly monotonically from 0 to 1.

    Thus if we take f to be a function which equals zero at every point where G has derivative zero, but f=1 elsewhere, then the integral of f is zero on [a,b], but the antiderivative G of f is not constant, hence does not differ from F by a constant. I.e. the mean value theorem fails for functions like G which are only differentiable almost everywhere, and hence the second FTC, which is essentially the mean value theorem, also fails in this form.

    However, we can recover a version of the mean value theorm, hence a second FTC by introducing the concept of Lipschitz continuity. G is Lipschitz continuous on [a,b] if there is some constant K such that for all x,y in [a,b], we always have |G(y)-G(x)| less than or equal to K|y-x|.

    1)B If f is Riemann integrable on [a,b], then f is continuous almost everywhere, hence the function F(x) = integral of f on [a,x], is differentiable with derivative equal to f almost everywhere. Moreover the integral function F is Lipschitz continuous on [a,b].

    2)B If f is Riemann integrable on [a,b], and G is any Lipschitz continuous function on [a,b] which is differentiable with derivative equal to f at those points where f is continuous, then G does differ from the integral F of f by a constant, i.e. G(x)-G(a) = F(x).


    The proof of this version follows from a stronger mean value theorem that shows that a Lipschitz continuous function with derivative zero a.e. is constant. you might try proving this using compactness.

    These theorems reach their final form for Lebesgue integrable fucntions using the still subtler concept of “absolute continuity” in place of Lipschitz continuity.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook