This page is a sub-page of our page on Limits.
///////
Related KMR pages:
• Differentiation and Affine Approximation in One Real Variable
• Taylor expansion in One Real Variable
///////
Other relevant sources of information:
• Big-Ordo notation
• Little-ordo notation
///////
Big-Ordo and little-ordo
The symbols \, O \, = “big Ordo” and \, o \, = “little ordo” describe how fast functions are decreasing towards zero when their independent variable is approaching some fixed point where they take the value zero.
Let \, f : \mathbb{R} \rightarrow \mathbb{R} \, be a function
and let its independent variable \, x approach a fixed point \, a \in \mathbb{R} where \, f(a) = 0 .
Definition: \, O_a(f) \stackrel {\mathrm{def}}{=} \{ g : \mathbb{R} \rightarrow \mathbb{R} \} \, such that, whenever \, x \, is close enough to \, a ,
we have \, | \,\dfrac{g(x)}{f(x)} \, | < M \} \, for some fixed, positive constant \, M \in \mathbb{R} .
Definition: \, o_a(f) \stackrel {\mathrm{def}}{=} \{ g : \mathbb{R} \rightarrow \mathbb{R} \} \, such that, whenever \, x \, is close enough to \, a ,
we have \, | \,\dfrac{g(x)}{f(x)} \, | < M \} \, for every fixed, positive constant \, M \in \mathbb{R} .
Intuitively, this means that:
\, O_a(f) \, denotes the set of functions that decrease at least as fast as \, f \,
when \, x \, approaches \, a ,
while
\, o_a(f) \, denotes the set of functions that decrease faster than \, f \,
when \, x \, approaches \, a .
NOTATION: When the point that is approached is clear from the context,
we write \, O(f) \, and \, o(f) \, instead of \, O_a(f) \, and \, o_a(f) .
///////
Rules of computation for \, O(f) \, :
1) \, f \cdot O(g) = O(f \cdot g) \,
2) \, O(f) \cdot O(g) = O(f \cdot g) \,
3) \, O(O(f)) = O(f) \,
In words:
1) A function \, f \,
times any function that decreases at least as fast as the function \, g \,
must decrease at least as fast as the function \, f \cdot g .
2) Any function that decreases at least as fast as the function \, f \,
times any function that decreases at least as fast as the function \, g \,
must decrease at least as fast as the function \, f \cdot g .
3) Any function that decreases at least as fast as
any function that decreases at least as fast as the function \, f \,
must decrease at least as fast as the function \, f .
///////
Rules of computation for \, o(f) \, :
1) \, f \cdot o(g) = o(f \cdot g) \,
2) \, o(f) \cdot o(g) = o(f \cdot g) \,
3) \, o(o(f)) = 0 \,
In words:
1) A function \, f \,
times any function that decreases faster than the function \, g \,
must decrease faster than the function \, f \cdot g .
2) Any function that decreases faster than the function \, f \,
times any function that decreases faster than the function \, g \,
must decrease faster than the function \, f \cdot g .
3) Any function that decreases faster than
any function that decreases faster than the function \, f \,
must be identically zero in some neighborhood of the point that is approached.
///////
Example:
The interactive simulation that created this movie
The blue line intersects the red parabola tangentially at the point \, P .
The green line intersects the red parabola transversally at the point \, P .
The blue point \, W \, approaches \, P \, along the blue line,
which intersects the parabola tangentially at \, P .
The black point \, Q \, approaches \, P \, along the parabola itself.
The green point \, R \, approaches \, P \, along the green line,
which intersects the parabola transversally at \, P .
The horizontal distance between the green and the red vertical lines is \, ∆ \, .
TAKE AWAY MESSAGE:
\, | \, QW \, | = o(∆)
\, | \, RW \, | = O(∆)
In general:
Let the parabola be replaced by any smooth curve
and fix a point \, P \, on this curve. Then the following is true:
Any point \, Q \, that approaches \, P \, along a curve
that intersects the given curve tangentially at the point \, P \,
behaves like the point \, Q \, in the movie, i.e., we have \, | \, QW \, | = o(∆) .
Any point \, R \, that approaches \, P \, along a curve
that intersects the given curve transversally (= non-tangentially) at the point \, P \,
behaves like the point \, R \, in the movie, i.e., we have \, | \, RW \, | = O(∆) .