Difference between revisions of "Implicit Differentiation"

From Grad Wiki
Jump to navigation Jump to search
Line 1: Line 1:
  
 
== Background ==
 
== Background ==
So far, you may only have differentiated functions written in the form
+
 
<math>y=f\left(x\right)</math>. But some functions are better described by an
+
So far, you may only have differentiated functions written in the form <math style="vertical-align: -5px">y=f(x)</math>. But some functions are better described by an equation involving <math>x</math> and <math style="vertical-align: -4px">y</math>. For example, <math style="vertical-align: -5px">x^{2}+y^{2}=16</math> describes the graph of a circle with center <math>\left(0,0\right)</math> and radius 4, and is really the graph of two functions: <math style="vertical-align: -5px">y=\pm\sqrt{16-x^{2}}</math>.
equation involving <math>x</math> and <math>y</math>. For example, <math>x^{2}+y^{2}=16</math> describes
 
the graph of a circle with center <math>\left(0,0\right)</math> and radius 4,
 
and is really the graph of two functions: <math>y=\pm\sqrt{16-x^{2}}</math>.
 
 
   
 
   
Sometimes, functions described by equations in <math>x</math> and <math>y</math> are too hard
+
Sometimes, functions described by equations in <math style="vertical-align: 0px">x</math> and <math style="vertical-align: -5px">y</math> are too hard to solve for <math style="vertical-align: -5px">y</math>, for example <math style="vertical-align: -5px">x^{3}+y^{3}=6xy</math>. This equation really describes 3 different functions of x, whose graph is the curve:
to solve for <math>y</math>, for example <math>x^{3}+y^{3}=6xy</math>. This equation really describes 3 different functions of x, whose graph
+
 
is the curve:
+
We want to find derivatives of these functions without having to solve for <math style="vertical-align: -5px">y</math> explicitly. We do this by implicit differentiation. The process is to take the derivative of both sides of the given equation with respect to <math>x</math>, and then do some algebra steps to solve for <math style="vertical-align: -5px">y'</math> (or <math>\dfrac{dy}{dx}</math> if you prefer), keeping in mind that <math style="vertical-align: -5px">y</math> is a function of <math>x</math> throughout the equation.
 +
 
 +
 
 +
== Warm-up exercises ==
 +
 
 +
Given that <math style="vertical-align: -5px">y</math> is a function of <math>x</math>, find the derivative of the
 +
following functions with respect to <math>x</math>.
 +
 
 +
1. <math style="vertical-align: -5px">y^{2}</math>
 +
 
 +
Solution: <math style="vertical-align: -5px">2yy'</math>
 +
 
 +
Reason: Think <math style="vertical-align: -5px">y=f(x)</math> and view it as <math style="vertical-align: -5px">(f(x))^{2}</math> to see that the
 +
derivative is <math style="vertical-align: -5px">2f(x)f'(x)</math> by the chain rule, but write it as <math style="vertical-align: -5px">2yy'</math>.
 +
 
 +
2. <math style="vertical-align: -5px">xy</math>
 +
 
 +
Solution: <math style="vertical-align: -5px">xy'+y</math>
 +
 
 +
Reason: <math>x</math> and <math style="vertical-align: -5px">y</math> are both functions of <math>x</math>, and they are being
 +
multiplied together, so the product rule says it's <math style="vertical-align: -5px">x\cdot y'+y\cdot1</math>.
 +
 
 +
3. <math style="vertical-align: -5px">\cos y</math>
 +
 
 +
Solution: <math style="vertical-align: -5px">-y'\sin y</math>
 +
 
 +
Reason: The function <math style="vertical-align: -5px">y</math> is inside of the cosine function, so the
 +
chain rule gives <math style="vertical-align: -5px">(-\sin y)\cdot y'</math>.
 +
 
 +
4. <math style="vertical-align: -5px">\sqrt{x+y}</math>
  
We want to find derivatives of these functions without having to solve
+
Solution: <math style="vertical-align: -5px">\frac{1+y'}{2\sqrt{x+y}}</math>
for <math>y</math> explicitly. We can do this by implicit differentiation,
 
in which we take the derivative of both sides of our equation with respect
 
to <math>x</math>, and do some algebra steps to solve for <math>y'</math> (or <math>\dfrac{dy}{dx}</math>
 
if you prefer), keeping in mind that <math>y</math> is a function of <math>x</math> throughout
 
the equation.
 
  
 +
Reason: Write it as <math style="vertical-align: -5px">(x+y)^{\frac{1}{2}}</math>, and use the chain rule
 +
to get <math style="vertical-align: -15px">\frac{1}{2}\left(x+y\right)^{-\frac{1}{2}}\cdot\left(1+y'\right)</math>,
 +
then simplify.
  
 
== Example 1 ==
 
== Example 1 ==

Revision as of 23:00, 17 November 2015

Background

So far, you may only have differentiated functions written in the form . But some functions are better described by an equation involving and . For example, describes the graph of a circle with center and radius 4, and is really the graph of two functions: .

Sometimes, functions described by equations in and are too hard to solve for , for example . This equation really describes 3 different functions of x, whose graph is the curve:

We want to find derivatives of these functions without having to solve for explicitly. We do this by implicit differentiation. The process is to take the derivative of both sides of the given equation with respect to , and then do some algebra steps to solve for (or if you prefer), keeping in mind that is a function of throughout the equation.


Warm-up exercises

Given that is a function of , find the derivative of the following functions with respect to .

1.

Solution:

Reason: Think and view it as to see that the derivative is by the chain rule, but write it as .

2.

Solution:

Reason: and are both functions of , and they are being multiplied together, so the product rule says it's .

3.

Solution:

Reason: The function is inside of the cosine function, so the chain rule gives .

4.

Solution:

Reason: Write it as , and use the chain rule to get , then simplify.

Example 1

Find if .

[Think and momentarily view the equation as to realize that the term requires the chain rule and the term needs the product rule when differentiating, while the derivative of 8 is just 0.]

Then we get


Example 2

Find the equation of the tangent line to at the point .

We first compute by implicit differentiation. Note the derivative of the right side requires the quotient rule.


At the point we have and . Plugging these into our equation for gives


This means the slope of the tangent line at is , and a point on this line is . Using the point-slope form of a line, we have


Example 3

Find if .

Use implicit differentiation to find first:


Now is just the derivative of with respect to (remember ). This will require the chain rule. Note that we already found the derivative of to be . So


But we mustn't leave in our final answer. So, plug back in to get


as our final answer.