Finding maxima, minima, or saddle points of a function

multivariable-calculuspartial derivative

I want to find the local maximum, local minimum, or saddle points of this function:

$$f(x,y)=y(e^x-1)=ye^x-y$$

$f_{x}=e^xy$

$f_{y}=e^x-1$

$f_{xx}=e^xy$

$f_{xy}=e^x$

$f_{yy}=0$

$$f_{y}=0=e^x-1\rightarrow x=0$$
$$f_{x}=0 = e^xy\rightarrow y=0$$

This means there is a critical point at (0,0).

Using the second derivative test to check whether this is a local minimum, local maximum, or neither:

$$D(0,0)=f_{xx}(0,0)\cdot f_{yy}(0,0)-[f_{xy}(0,0)]^2$$
$$D(0,0)=0\cdot 0-1 = -1$$

Since $D(0,0) < 0$, the point $(0,0)$ is neither a local maximum or local minimum, so it is a saddle point.

Have I done this correctly?

Best Answer

It is correct. For |x|<<1 and |y|<<1, your function behaves just like f(x,y)=x*y which definitively has a (the classical!) saddle point S(0,0). The saddle though would not be too comfortable to sit on, due to its flattening out for values x < 0.