Partial differentiation

Consider
Partial differentiation cont..
• Now consider
𝜕𝑓 𝑓 𝑥 + ℎ, 𝑦 − 𝑓 𝑥, 𝑦
= 𝑙𝑖𝑚ℎ→0
𝜕𝑥 ℎ

𝜕𝑓 𝑓 𝑥, 𝑦 + ℎ − 𝑓 𝑥, 𝑦
= 𝑙𝑖𝑚ℎ→0
𝜕𝑦 ℎ
Partial differentiation cont..

Consider
• 𝑧 = 𝑥𝑦 2 + 𝑦𝑥 3
• 𝑧 = 𝑓 𝑥, 𝑦
𝜕𝑧
• = 𝑦 2 + 3𝑦𝑥 2
𝜕𝑥
𝜕𝑧
• = 2𝑥𝑦 + 𝑥 3
𝜕𝑦

With respect to three-dimensional graphs, you can picture the

𝜕𝑧 𝜕𝑧
partial derivatives, and which entail how the function will
𝜕𝑥 𝜕𝑦
change as one variable change while holding other to be constant.
• The swirl-d symbol, 𝜕, often called ‘del’ is used to
distinguish partial derivative from ordinary single variable
derivative.
• The reason for partial derivative is that when the input of
the function is made up of multiple variables, we want to
see how the function changes as we let one variable
change while keeping other to be constant.
Partial differentiation cont..
Partial differentiation cont..
Partial differentiation cont..
Note:
Examples
1. Determine the partial derivative of the function:
a) 𝑓 𝑥, 𝑦 = 3𝑥 + 4𝑦
b) 𝑓 𝑥, 𝑦 = 𝑥 2 𝑦 + 𝑠𝑖𝑛 𝑥 + 𝑐𝑜𝑠 𝑦 .
2. Given the function, 𝑓 𝑥, 𝑦, 𝑧 = 𝑥𝑐𝑜𝑠 𝑧 + 𝑥 2 𝑦 3 𝑒 𝑧 . Find:-
a) 𝜕𝑓Τ𝜕𝑥
b) 𝜕𝑓Τ𝜕𝑦
c) 𝜕𝑓Τ𝜕𝑧
Higher derivatives
Higher derivatives
Theorem

Assume that 𝑓 and all its partial derivatives 𝑓𝑥 and 𝑓𝑦 are continuous,
and that 𝑥 = 𝑥 𝑡 and y = 𝑦 𝑡 are themselves differentiable
functions of 𝑡. Let 𝐹 𝑡 = 𝑓 𝑥 𝑡 , 𝑦 𝑡 Then 𝐹 is differentiable and

𝑑𝐹 𝜕𝑓 𝑑𝑥 𝜕𝑓 𝑑𝑦
= . + .
𝑑𝑡 𝜕𝑥 𝑑𝑡 𝜕𝑦 𝑑𝑡
 𝑑𝑓 𝜋
• Let 𝑓 𝑥, 𝑦 = 𝑥𝑦, 𝑥 = 𝑐𝑜𝑠 𝑡 , y = sin 𝑡 . Compute when 𝑡 = .
𝑑𝑡 2
Example
Assume that 𝑓 𝑢, 𝑣, 𝑤 has continuous partial derivatives,
and that 𝑢 = 𝑥 − 𝑦; 𝑣 = 𝑦 − 𝑧; 𝑤 = 𝑧 − 𝑥
Let
𝐹 𝑥, 𝑦, 𝑧 = 𝑓 𝑢 𝑥, 𝑦, 𝑧 , 𝑣 𝑥, 𝑦, 𝑧 , 𝑤 𝑥, 𝑦, 𝑧
Show that
𝜕𝐹 𝜕𝐹 𝜕𝐹
+ + =0
𝜕𝑥 𝜕𝑦 𝜕𝑧
Example
• Suppose that 𝑓 𝑥, 𝑦 is written in terms of 𝑢 and 𝑣 where
𝑥 = 𝑢 + log 𝑣 and y= 𝑢 − log 𝑣 . Show that, with the usual
convention
𝜕2 𝑓 𝜕2 𝑓 𝜕2 𝑓 𝜕2 𝑓
a. 𝜕𝑢 2 = 2+2
𝜕𝑥 𝜕𝑥𝜕𝑦
+ 2
𝜕𝑦
2 𝜕2 𝑓 𝜕2 𝑓 𝜕2 𝑓
2𝜕 𝑓 𝜕𝑓 𝜕𝑓
b. 𝑣
𝜕𝑣 2 = − + 2+
𝜕𝑦 𝜕𝑥 𝜕𝑥
2
𝜕𝑥𝜕𝑦
+
𝜕𝑦 2
Minima and Maxima
Minima and Maxima
Definition
Say that 𝑓 𝑥, 𝑦 has a critical point at 𝑎, 𝑏 iff:-
𝜕𝑓 𝜕𝑓
𝑎, 𝑏 = 𝑎, 𝑏 = 0
𝜕𝑥 𝜕𝑦
It is clear by comparison with the single variable result, that a
necessary condition that if have a local extremum at (a, b) is
that it have a critical point there, although that is not a sufficient
condition. We refer to this as the first derivative test.
We can get more information by looking at the second
derivative Theorem (Second Derivative Test). Assume that
(a; b) is a critical point for f, then:-