Not What You Meant? There are 54 definitions for Jacob. Also try: Jacobite or Jacobi matrix.

Jacobian

About 5 pages (1,575 words)

Know this topic well? Help others and get FREE products!

For the French Revolution faction, see Jacobin. For the followers of James II of England and VII of Scotland, see Jacobitism. For other uses see Jacobean.

In vector calculus, the Jacobian is shorthand for either the Jacobian matrix or its determinant, the Jacobian determinant. In algebraic geometry the Jacobian of a curve means the Jacobian variety: a group variety associated to the curve, in which the curve can be embedded. These concepts are all named after the mathematician Carl Gustav Jacobi. The term "Jacobian" is normally pronounced [jaˈkobiən], but can also be pronounced [ʤəˈkobiən].

1 Jacobian matrix
2 Jacobian determinant
- 2.1 Example
- 2.2 Uses
3 See also
4 References
5 External links

Jacobian matrix

The Jacobian matrix is the matrix of all first-order partial derivatives of a vector-valued function. If a function is differentiable at a point, its derivative is given in coordinates by the Jacobian, but a function need not be differentiable for the Jacobian to be defined: only the partial derivatives need exist. Its importance lies in the fact that it represents the best linear approximation to a differentiable function near a given point. In this sense, the Jacobian is the derivative of a multivariate function. For n > 1, the derivative of a numerical function must be matrix-valued, or a partial derivative. Suppose F : Rⁿ → R^m is a function from Euclidean n-space to Euclidean m-space. Such a function is given by m real-valued component functions, y₁(x₁,...,x_n), ..., y_m(x₁,...,x_n). The partial derivatives of all these functions (if they exist) can be organized in an m-by-n matrix, the Jacobian matrix of F, as follows:

<math>\begin{bmatrix} \frac{\partial y_1}{\partial x_1} & \cdots & \frac{\partial y_1}{\partial x_n} \\ \vdots & \ddots & \vdots \\ \frac{\partial y_m}{\partial x_1} & \cdots & \frac{\partial y_m}{\partial x_n} \end{bmatrix}. </math>

This matrix is denoted by

<math>J_F(x_1,\ldots,x_n)</math> or by <math>\frac{\partial(y_1,\ldots,y_m)}{\partial(x_1,\ldots,x_n)}.</math>

The ith row of this matrix is given by the transpose of the gradient of the function y_i for i = 1,...,m. If p is a point in Rⁿ and F is differentiable at p, then its derivative is given by J_F(p) (and this is the easiest way to compute the derivative). In this case, the linear map described by J_F(p) is the best linear approximation of F near the point p, in the sense that

<math>F(\mathbf{x}) = F(\mathbf{p}) + J_F(\mathbf{p})(\mathbf{x}-\mathbf{p}) + o(|\mathbf{x}-\mathbf{p}|)</math>

for x close to p and where o(...) is the little o-notation. Note that the Jacobian of the gradient is the Hessian matrix. If the components of F are arranged into a column vector

<math>\mathbf{y} = (y_1, \dots, y_m)</math>

the Jacobian may be represented as an outer product between the del operator and <math>\mathbf{y}</math>:

<math>\frac{\partial(y_1,\ldots,y_m)}{\partial(x_1,\ldots,x_n)} = \nabla \otimes \mathbf{y}</math>

where the outer product symbol is often left out, it being understood that the gradient of a column vector is contextually a matrix.

Differentiable versus partial derivatives

The Jacobian matrix requires a coordinate system: there is no coordinate-free notion of partial derivative, only directional derivative. In the absence of a coordinate system, instead of "matrix of first partials", one defines the "best linear approximation", which is a stronger concept, as demonstrated below. This notion is defined for a general map between Banach spaces, and is called the Fréchet derivative. For the Jacobian to be defined at a point p, one only needs the partial derivatives to be defined, which is weaker than for the function to be differentiable at that point: partial derivatives require that the function has a linear approximation along the coordinate axes, while differentiability requires the function to have a linear approximation that works in every direction at once.

$The function <math>f(r,\theta) = r \sin (2\theta)</math>. Its Jacobian at the origin in the <math>x,y</math> coordinate system is all zero, which is a perfect linear approximation in the coordinate directions, but a bad approximation in all other directions.$

The function <math>f(r,\theta) = r \sin (2\theta)</math>. Its Jacobian at the origin in the <math>x,y</math> coordinate system is all zero, which is a perfect linear approximation in the coordinate directions, but a bad approximation in all other directions.

For example, the function given in polar coordinates by

<math>f(r,\theta) = r \sin (2\theta)</math>

is identically zero on the x and y axes, and hence has a well-defined Jacobian matrix at the origin (with respect to the <math>(x,y)</math> coordinate system), which is zero, but the zero matrix does not give a good linear approximation (since it does not agree with the directional derivatives, given by <math>\lim_{h\to 0^+} [f(hv)-f(0)]/h</math>; for instance, in the <math>(1,1)</math> direction (45°) this equals <math>\sin 45^\circ=\sqrt{2}/2</math>, which is not zero), and the function is not differentiable at the origin—if it were, the matrix for the derivative would have to be the Jacobian matrix, since that is the unique best linear approximation on the axes.

Examples

The transformation from spherical coordinates to Cartesian coordinates is given by the function F : R × [0,π] × [0,2π] → R³ with components:

<math> x_1 = r \sin\phi \cos\theta \,</math>

<math> x_2 = r \sin\phi \sin\theta \,</math>

The Jacobian matrix for this coordinate change is

<math>J_F(r,\phi,\theta) =\begin{bmatrix}

\frac{\partial x_1}{\partial r} & \frac{\partial x_1}{\partial \phi} & \frac{\partial x_1}{\partial \theta} \\[3pt] \frac{\partial x_2}{\partial r} & \frac{\partial x_2}{\partial \phi} & \frac{\partial x_2}{\partial \theta} \\[3pt] \frac{\partial x_3}{\partial r} & \frac{\partial x_3}{\partial \phi} & \frac{\partial x_3}{\partial \theta} \\ \end{bmatrix}=\begin{bmatrix} \sin\phi \cos\theta & r \cos\phi \cos\theta & -r \sin\phi \sin\theta \\ \sin\phi \sin\theta & r \cos\phi \sin\theta & r \sin\phi \cos\theta \\ \cos\phi & -r \sin\phi & 0 \end{bmatrix}. </math> The Jacobian matrix of the function F : R³ → R⁴ with components

<math>J_F(x_1,x_2,x_3) =\begin{bmatrix}

\frac{\partial y_1}{\partial x_1} & \frac{\partial y_1}{\partial x_2} & \frac{\partial y_1}{\partial x_3} \\[3pt] \frac{\partial y_2}{\partial x_1} & \frac{\partial y_2}{\partial x_2} & \frac{\partial y_2}{\partial x_3} \\[3pt] \frac{\partial y_3}{\partial x_1} & \frac{\partial y_3}{\partial x_2} & \frac{\partial y_3}{\partial x_3} \\[3pt] \frac{\partial y_4}{\partial x_1} & \frac{\partial y_4}{\partial x_2} & \frac{\partial y_4}{\partial x_3} \\ \end{bmatrix}=\begin{bmatrix} 1 & 0 & 0 \\ 0 & 0 & 5 \\ 0 & 8x_2 & -2 \\ x_3\cos(x_1) & 0 & \sin(x_1) \end{bmatrix}. </math> This example shows that the Jacobian need not be a square matrix.

In dynamical systems

Consider a dynamical system of the form x' = F(x), with F : Rⁿ → Rⁿ. If F(x₀) = 0, then x₀ is a stationary point. The behavior of the system near a stationary point can often be determined by the eigenvalues of J_F(x₀), the Jacobian of F at the stationary point.^[1]

Jacobian determinant

If m = n, then F is a function from n-space to n-space and the Jacobian matrix is a square matrix. We can then form its determinant, known as the Jacobian determinant. The Jacobian determinant is also called the "Jacobian" in some sources. The Jacobian determinant at a given point gives important information about the behavior of F near that point. For instance, the continuously differentiable function F is invertible near p if the Jacobian determinant at p is non-zero. This is the inverse function theorem. Furthermore, if the Jacobian determinant at p is positive, then F preserves orientation near p; if it is negative, F reverses orientation. The absolute value of the Jacobian determinant at p gives us the factor by which the function F expands or shrinks volumes near p; this is why it occurs in the general substitution rule.

Example

The Jacobian determinant of the function F : R³ → R³ with components

<math>\begin{vmatrix} 0 & 5 & 0 \\ 8x_1 & -2x_3\cos(x_2 x_3) & -2x_2\cos(x_2 x_3) \\ 0 & x_3 & x_2 \end{vmatrix}=-8x_1\cdot\begin{vmatrix} 5 & 0\\ x_3&x_2\end{vmatrix}=-40x_1 x_2.</math>

From this we see that F reverses orientation near those points where x₁ and x₂ have the same sign; the function is locally invertible everywhere except near points where x₁ = 0 or x₂ = 0. If you start with a tiny object around the point (1,1,1) and apply F to that object, you will get an object set with about 40 times the volume of the original one.

Uses

The Jacobian determinant is used when making a change of variables when integrating a function over its domain. To accommodate for the change of coordinates the Jacobian determinant arises as a multiplicative factor within the integral. Normally it is required that the change of coordinates is done in a manner which maintains an injectivity between the coordinates that determine the domain. The Jacobian determinant, as a result, is usually well defined.