1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Local approximation

  1. Oct 22, 2004 #1
    g'(3)= -2


    I got this question wrong on a test, were was my mistake?
  2. jcsd
  3. Oct 22, 2004 #2
    The primes on the function indicate the tangents at a point.

    So f(x) is the value of the function at the point x = x and f'(x) is the value of the tangent at the point x = x. You were asked to find the tangent at an intermediate point given the tangents at the two extremeties of a subinterval [2,3]. You claim (in your solution) that the derivative at any intermediate point equals the ratio of the difference in derivatives at the end points of the subinterval to the length of the subinterval. Now think about this: your expression does not involve the point x = 2.5 anywhere. So it would be just as well if I could replace 2.5 by 2.75 and use your expression...which would mean that the tangents (or slopes) of the curve at both x = 2.5 and x = 2.75 are equal! Is that so? Not unless the function is constant in the subinterval [2.5, 2.75] but that again is not something you can assume.

    Also what is msec = mtan?
  4. Oct 23, 2004 #3
    the slope of secant = the slope of the tangent
  5. Oct 23, 2004 #4
    Is that what you call a local approximation? Hmm...why do you think its wrong then?
  6. Oct 23, 2004 #5


    User Avatar
    Staff Emeritus
    Science Advisor

    The fact that these are derivatives is irrelevant. We are given that a function, f (which happens to be g' in this problem) at x= 2 and x= 3 and asked to "approximate" f(2.5). Given g'(2) and g'(3) there is NO way of saying for sure what g'(2.5) is but there are an infinite number of different "approximations" which may or may not be accurate.

    The SIMPLEST approximation, given two points, is "linear interpolation". Since f(2)= 1 and f(3)= -2, the linear interpolation is just the number halfway between those:
    (1+ (-2))/2= -1/2.

    UrbanXrisis: IF you were given g(2)= 1 and g(3)= -2 (Not g' ) THEN the simplest estimate for g' anywhere between 2 and 3 would be the slope of the straight line:
    (-2-1)/(2-1)= -3 but that was NOT what you were given!
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Local approximation
  1. Local Maxima (Replies: 11)

  2. Local linearization (Replies: 6)

  3. Local form? (Replies: 1)

  4. Approximation of SHM (Replies: 6)