62. Material en revisión: miércoles 23 de octubre

Por Mariana Perez

Teorema

Sea $ f : \mathcal{U} \subseteq \mathbb{R}^2 \rightarrow \mathbb{R}$ tal que existen las segundas derivadas parciales.

Si son continuas, entonces el polinomio de Taylor de 2° grado es el único $p (x, y)$ tal que

$$\lim\limits_{(h,k) \to (0,0)} \dfrac{f (x_0 + h, y_0 + k) \, – \, p (x_0 + h, y_0 + k)}{\|(h,k)\|^2} = 0 $$

Observación: si NO son continuas la demostración no sería válida.

Demostración:

$p (x_0 + h, y_0 + k) = Ah^2 + B hk + C k^2 + Dh + Ek + F$ para algunas $A, B, C, D, E, F \in \mathbb{R}$ constantes.

$ F = f (h, k)$ ya que $ \lim\limits_{(h,k) \to (0,0) } f (x_0 + h, y_0 + k) \, – \, p (x_0 + h, y_0 + k) = f (x_0 , y_0 ) \, – \, F = 0$

Tenemos que $D = \dfrac{\partial f}{\partial x} (x_0, y_0)$, y $ E = \dfrac{\partial f}{\partial y} (x_0, y_0)$ porque

$\lim\limits_{(h,k) \to (0,0)} \dfrac{\big| f (x_0 + h, y_0 + k) \, – \, p (x_0 + h, y_0 + k) \big| }{\| (h,k)\|} = 0$

En particular, si tomamos el límite con puntos de la forma $(h, 0)$ y $(0, k)$ entonces

$\lim\limits_{h \to 0} \dfrac{\big| f (x_0 + h, y_0 ) \, – \, p (x_0 + h, y_0 ) \big| }{\big| (h,k) \big|} = 0$

$\lim\limits_{h \to 0} \dfrac{\big| f (x_0 + h, y_0 ) \, – \, f (x_0 , y_0 ) \, – \, Dh \, – \, Ah^2 \big| }{\big| h \big|} = 0$

$\iff$

$\lim\limits_{h \to 0} \dfrac{ f (x_0 + h, y_0 ) \, – \, f (x_0 , y_0 )}{h} \, – \, D \, – \, \cancel{A\lim\limits_{h \to 0} |h|}^0 = 0$

$\iff$

$\lim\limits_{h \to 0} \dfrac{ f (x_0 + h, y_0 ) \, – \, f (x_0 , y_0 )}{h} = D$

$\iff$

$D = \dfrac{\partial f}{\partial x} (x_0, y_0)$

Análogamente $E = \dfrac{\partial f}{\partial y} (x_0, y_0)$

Ahora bien,

$$\lim\limits_{(h,k) \to (0,0)} \dfrac{f (x_0 + h, y_0 + k) \, – \, p (x_0 + h, y_0 + k)}{\|(h,k)\|^2} = \lim\limits_{(h,k) \to (0,0)} \dfrac{\big| f (x_0 + h, y_0 ) \, – \, f (x_0 , y_0 ) \, – \, \dfrac{\partial f}{\partial x} (x_0, y_0) h \, – \, \dfrac{\partial f}{\partial y} (x_0, y_0) k \, – \, Ah^2 \, – \, 2B hk \, – \, C k^2 \big| }{\big\|( h, k) \big\|^2} $$

En particular, si tomamos puntos de la forma $(h, 0)$

$\lim\limits_{h \to 0} \dfrac{\big| f (x_0 + h, y_0 ) \, – \, f (x_0 , y_0 ) \, – \, \dfrac{\partial f}{\partial x} (x_0, y_0) h \, – \, Ah^2 \big| }{\big| h \big|^2} = 0 $

$\iff$

$\lim\limits_{h \to 0} \dfrac{ f (x_0 + h, y_0 ) \, – \, f (x_0 , y_0 ) \, – \, \dfrac{\partial f}{\partial x} (x_0, y_0) h \, – \, Ah^2 }{ h^2 } = 0 $

$\iff$

$\lim\limits_{h \to 0} \dfrac{ f (x_0 + h, y_0 ) \, – \, f (x_0 , y_0 ) \, – \, \dfrac{\partial f}{\partial x} (x_0, y_0) h }{ h^2 } = A $

Aplicando la regla de L’Hôpital

$\lim\limits_{h \to 0} \dfrac{ f (x_0 + h, y_0 ) \, – \, f (x_0 , y_0 ) \, – \, \dfrac{\partial f}{\partial x} (x_0, y_0) h }{ h^2 } = \lim\limits_{h \to 0} \dfrac{ \dfrac{\partial f}{\partial x} (x_0 + h, y_0 ) \, – \, \dfrac{\partial f}{\partial y} (x_0 , y_0 ) }{ 2h } = \lim_{h \to 0} \frac{G (h)}{2h} $ . . . (*)

Aplicaremos L’Hôpital por segunda vez a la expresión anterior (*), entonces tenemos

$\lim\limits_{h \to 0 } \dfrac{G’ (h)}{2}$

Necesitamos que $\lim\limits_{h \to 0 } G (h) = 0 $ , es decir ,

$ \lim\limits_{h \to 0} \dfrac{\partial f}{\partial x} (x_0 + h, y_0 ) \, – \, \dfrac{\partial f}{\partial y} (x_0 , y_0 ) = 0$

Esto lo garantizamos si $\dfrac{\partial f}{\partial x} (x, y) $ es continua en $(x_0, y_0)$.

Luego

$G’ (h) = \dfrac{\partial^2 f}{\partial x^2} (x_0 + h, y_0)$ por continuidad, entonces cuando $h \rightarrow 0 , G’ (0) = \dfrac{\partial^2 f}{\partial x^2} (x_0 , y_0)$

Por lo tanto, $A = \dfrac{1}{2} \dfrac{\partial^2 f}{\partial x^2} (x_0, y_0)$

Análogamente usando trayectorias $x_0, y_0 + k)$, cuando $ k \rightarrow 0$ tenemos que $ C = \dfrac{1}{2} \dfrac{\partial^2 f}{\partial y^2} (x_0, y_0)$

Reuniendo la información anterior , tenemos que

$\lim\limits_{(h,k) \to (0,0)} \dfrac{\big| f (x_0 + h, y_0 + k ) \, – \, f (x_0 , y_0 ) \, – \, \dfrac{\partial f}{\partial x} (x_0, y_0) h \, – \, \dfrac{\partial f}{\partial y} (x_0, y_0) k \, – \, \dfrac{1}{2} \dfrac{\partial^2 f}{\partial x^2} (x_0, y_0) h^2 \, – \, 2B hk \, – \, \dfrac{1}{2} \dfrac{\partial^2 f}{\partial y^2} (x_0, y_0) k^2 \big| }{\big\|( h, k) \big\|^2} = 0 $

En particular cuando tomamos $h = k$

$\lim\limits_{(h,h) \to (0,0)} \dfrac{\big| f (x_0 + h, y_0 + h ) \, – \, f (x_0 , y_0 ) \, – \, \dfrac{\partial f}{\partial x} (x_0, y_0) h \, – \, \dfrac{\partial f}{\partial y} (x_0, y_0) h \, – \, \dfrac{1}{2} \dfrac{\partial^2 f}{\partial x^2} (x_0, y_0) h^2 \, – \, 2B h^2 \, – \, \dfrac{1}{2} \dfrac{\partial^2 f}{\partial y^2} (x_0, y_0) h^2 \big| }{ 2h^2} = 0 $

$\iff$

$\lim\limits_{ h \to 0} \dfrac{ f (x_0 + h, y_0 + h ) \, – \, f (x_0 , y_0 ) \, – \, h \Big( \dfrac{\partial f}{\partial x} (x_0, y_0) \, – \, \dfrac{\partial f}{\partial y} (x_0, y_0) \Big) \, – \, \dfrac{1}{2} h^2 \Big( \dfrac{\partial^2 f}{\partial x^2} (x_0, y_0) \, – \, \dfrac{\partial^2 f}{\partial y^2} (x_0, y_0) \Big) }{ 2h^2} = 2 B $

$\iff$ Aplicando L’Hôpital

$\lim\limits_{ h \to 0} \dfrac{ \dfrac{\partial f}{\partial x} (x_0 + h, y_0 + h ).1 \, – \, \dfrac{\partial f}{\partial y} (x_0 , y_0 ).1 \, – \, \dfrac{\partial f}{\partial x} (x_0, y_0) \, – \, \dfrac{\partial f}{\partial y} (x_0, y_0) h \, – \, \dfrac{\partial^2 f}{\partial y^2} (x_0, y_0) h }{ 4h} = B $

Aplicamos L’Hôpital nuevamente, entonces tenemos que:

$\lim\limits_{ h \to 0} \dfrac{ \dfrac{\partial^2 f}{\partial x^2} (x_0 + h, y_0 + h ) \, + \, \dfrac{\partial^2 f}{\partial y \; \partial x} (x_0 + h, y_0 + h ) \, +\, \dfrac{\partial^2 f}{\partial x \; \partial y} (x_0 + h, y_0 + h) \, + \, \dfrac{\partial^2 f}{\partial y^2} (x_0 + h, y_0 + h )\, – \, \dfrac{\partial^2 f}{\partial x^2} (x_0, y_0) \, – \, \dfrac{\partial^2 f}{\partial y^2} (x_0, y_0) h }{ 4} = B $

Si las segundas derivadas parciales son continuas, el último límite es igual a

$\dfrac{\dfrac{\partial^2 f}{\partial y \; \partial x} (x_0 , y_0 )}{4} = B$

y por tanto

$$B = \dfrac{1}{2} \dfrac{\partial^2 f}{\partial x \; \partial y } (x_0, y_0)$$

Decir que $p (x, y)$ es la mejor aproximación de 2° grado de $f (x, y)$ cerca del punto $(x_0 , y_0)$, es decir que

$$\lim\limits_{(h,k) \to (0,0)} \dfrac{\Big| f (x_0 + h, y_0 + k) \, – \, p (x_0 + h, y_0 + k) \Big|}{\Big\|(h,k) \Big\|^2} = 0 \; \; \; _{\blacksquare}$$

${}$

Regresando al ejemplo de derivadas parciales NO continuas. (puedes hacer click en el siguiente enlace para acceder a esa entrada)

https://blog.nekomath.com/?p=101015&preview=true

Dada la función

$\begin{equation*} f (x, y) = \begin{cases} \dfrac{xy ( x^2 \, – \, y^2) }{x^2 + y^2} & \text{ si } (x, y) \neq (0, 0) \\ {} \\ 0 & \text{ si } (x, y) = (0, 0) \end{cases} \end{equation*}$

¿Cuál será el polinomio de Taylor de 2° grado de $f$ alrededor del origen?

$f (0, 0) = 0$

$\dfrac{\partial f}{\partial x} (0, 0) = \lim\limits_{h \to 0} \dfrac{f (h, 0) \, – \, f (0, 0)}{h} = \lim\limits_{h \to 0} \dfrac{0}{h} = 0 $

Análogamente

$\dfrac{\partial f}{\partial y} (0, 0) = 0 $

$\dfrac{\partial^2 f}{\partial x^2} (0, 0) = \lim\limits_{h \to 0} \dfrac{\dfrac{\partial f}{\partial x} (h, 0) \, – \, \dfrac{\partial f}{\partial x} (0, 0)}{h} = \dfrac{0}{h} = 0 $

Análogamente

$\dfrac{\partial^2 f}{\partial y^2} (0, 0) = 0$

$\dfrac{\partial^2 f}{\partial x \partial y} (0, 0) = 1$

$\dfrac{\partial^2 f}{\partial y \; \partial x} (0, 0) = \,- \, 1$

Podemos examinar que pasa si tomamos un polinomio de la forma

$$p (x, y) = Ax^2 + Bxy + Cy^2 + Dx + Ey + F $$

$ F = 0$

$D = E = \nabla f (0, 0) = \vec{0} $ por lo tanto $ D = 0 = E $

Si $\lim\limits_{(h,k) \to (0,0)} \dfrac{\Big| f (h,k) \, – \, p (h,k) \Big|}{\Big\| (h,k) \Big\|^2} = 0$, entonces

Para $ k = 0$

$\lim\limits_{h \to 0} \dfrac{\Big| f (h,0) \, – \, p (h,0) \Big|}{h^2} = \lim\limits_{h \to 0} \dfrac{0 \, – \, Ah^2}{h} = A = 0$

Análogamente, para $h = 0$ tenemos que $C = 0$

luego

$\lim\limits_{(h,k) \to (0,0)} \dfrac{\Big| f (h,k) \, – \, 2Bhk \Big|}{\Big\| (h,k) \Big\|^2} = 0 $

En particular, para cuando $\lim\limits_{h \to 0} \dfrac{ f (h,h) \, – \, 2Bh^2}{2h^2} = 0$

$f (h, h) = 0$

$\lim\limits_{h \to 0} \dfrac{ \, – \, \cancel{2} B \cancel{h^2} }{\cancel{2} \cancel{h^2} } = \lim\limits_{h \to 0} \, – \, 2 B = 0$

Por lo tanto $ B = 0$

${}$

Sea $f (x, y) = ( y \, – \, 3x^2) ( y \, – \, x^2)$

¿Cómo son las curvas de nivel de $f$?

¿Cómo es la gráfica de $f$?

«Curva» de nivel 0

$ \Big\{ (x, y) \in \mathbb{R}^2 \Big| f (x, y) = 0 \} = f^{-1} ( 0 ) = \{ (x, y) \in \mathbb{R}^2 \Big| ( y \, – \, 3x^2) ( y \, – \, x^2) = 0 \Big\}$

Entonces

$y \, – \, 3x^2 = 0 \; \iff \; y = 3x^2$ o

$ y \, – \, x^2 = 0 \; \iff \; y = x^2 $

Si $ (x, y) $ cumple que $ x^2 < y < 3 x^2 $ entonces

$0 < y \, – \, x^2 $ pero $ y \, – \, 3 x^2 < 0 $ por lo tanto $ f (x, y) < 0 $

Si $ (x, y) $ cumple que $ y > 3 x^2 $ entonces $ f (x, y) > 0 $ ya que $ y > x^2$

61. Material en revisión: lunes 21 de octubre

Por Mariana Perez

En esta entrada pondremos en práctica lo aprendido anteriormente, analizando tres ejemplos.

Para cada una de las siguientes funciones buscamos:

(*) Indentificar los puntos críticos.

(*) Calcular el polinomio de Taylor de 2° grado alrededor de los puntos críticos y utilizarlo para saber si la función alcanza un máximo local, mínimo local o punto silla, si, el punto crítico es NO degenerado.

Ejemplo 1

Dada $f (x , y) = \ln ( 1 + x^2 + y^2 ) $

Analicemos cuando las derivadas parciales valen cero.

$\dfrac{\partial f}{\partial x} = \dfrac{1}{1+x^2+y^2} 2x = 0 \Rightarrow \dfrac{2x}{1+x^2+y^2} = 0 \Rightarrow x = 0$

$\dfrac{\partial f}{\partial y} = \dfrac{1}{1+x^2+y^2} 2y = 0 \Rightarrow \dfrac{2y}{1+x^2+y^2} = 0 \Rightarrow y = 0$

Por lo que el punto $(0, 0)$ es el único punto crítico.

El valor crítico correspondiente es

$f (0, 0) = \ln ( 1 + (0)^2 + (0)^2 ) = \ln (1) = 0$

Segundas derivadas parciales

$\dfrac{\partial^2 f}{\partial x^2} = \dfrac{2 (1 + x^2 + y^2) \, – \, 2x (2x)}{(1 + x^2 + y^2)^2} = \dfrac{2+2x^2+2y^2 \, – \, 4x^2}{(1 + x^2 + y^2)^2} $

Por lo tanto $\dfrac{\partial^2 f}{\partial x^2} = \dfrac{2+2y^2 \, – \, 2x^2}{(1 + x^2 + y^2)^2}$

Luego $f_{xx} (0, 0) = 2$

Análogamente

$\dfrac{\partial^2 f}{\partial y^2} = \dfrac{2 (1 + x^2 + y^2) \, – \, 2y (2y)}{(1 + x^2 + y^2)^2} = \dfrac{2+2x^2+2y^2 \, – \, 4y^2}{(1 + x^2 + y^2)^2} $

Por lo tanto $\dfrac{\partial^2 f}{\partial x^2} = \dfrac{2+2y^2 \, – \, 2y^2}{(1 + x^2 + y^2)^2}$

Luego $f_{yy} (0, 0) = 2$

Además

$\dfrac{\partial^2 f}{\partial y \, \partial x} = \dfrac{2 x(2 y) \, – \, 0}{(1 + x^2 + y^2)^2} = \dfrac{ 4 xy}{(1 + x^2 + y^2)^2} $

Luego $f_{xy} (0, 0) = 0$

Entonces el polinomio de Taylor de 2° grado de $f$ alrededor del punto $( 0, 0)$ es

$$p ( x, y) = \dfrac{1}{2} \big( 2 ( x^2 + y^2 ) \big) = x^2 + y^2 $$

EL punto crítico es NO degenerado, porque el determinante

$$\begin{vmatrix} f_{xx} & f_{xy} \\ \\ f_{xy} & f_{yy} \end{vmatrix} \neq 0$$

Por lo tanto, $f$ alcanza un valor mínimo en $ ( 0, 0).$

Observación:

En coordenadas polares

$$ f ( r, \theta ) = \ln ( 1 + r^2) $$

Las curvas de nivel son circunferencias.

(*) Curva de nivel 1

$\big\{ (x, y) \in \mathbb{R}^2 \big| \ln (1+x^2+y^2) = 1 \big\}$

$ 1+x^2+y^2 = e \; \; \iff \; \; x^2+y^2 = e \, – \, 1$ circunferencia de radio $ \sqrt{e \, – \, 1 \, }$

(*) Curva de nivel $\mathcal{c}$ ( con $\mathcal{c} > 1 $) es

$\big\{ (x, y) \in \mathbb{R}^2 \big| \ln (1+x^2+y^2) = \mathcal{c} \big\}$

$ 1+x^2+y^2 = e^{\mathcal{c}} \; \; \iff \; \; x^2+y^2 = e^{\mathcal{c}} \, – \, 1$ circunferencia de radio $ \sqrt{e^{\mathcal{c}} \, – \, 1 \, }$

Cerca de $ r = 0 $ , $ \ln ( 1 + r^2) $ se aproxima a su polinomio de Taylor.

$ f (r) = \ln (1+r^2)$

$ f (0) = 0$

$f’ (r) = \dfrac{2r}{1+r^2} $

$ f’ (0) = 0$

${f}^{\prime \prime} (r) = \dfrac{(1+r^2) 2 \, – \, 2r (2r) }{(1+r^2)^2} = \dfrac{2 + 2r^2 \, – \, 4r^2}{(1+r^2)^2} = \dfrac{ 2 \, – \, 2r^2 }{(1+r^2)^2} $

${f}^{\prime \prime} (0) = 2$

$ p (r) = \dfrac{1}{2} 2 r^2 = r^2$

Cerca de $ r = 0$ , $\ln (1+r^2) \approx r^2$

$\ln ( 1+x^2 + y^2) \approx x^2 + y^2 $ cuando $(x , y) $ está cerca del $ ( 0, 0) .$

Lejos de $ r = 0 $, $r$ es grande entonces, $ \ln (1+r^2) \approx \ln (r^2) = 2 \ln r $

IMAGEN

La gráfica de $f$ es $ \big\{ (x, y, z) \in \mathbb{R}^3 \Big| z = \ln (1 + x^2 + y^2) \big\} $ , una superficie de revolución girando la curva $ z = \ln (1 + x^2) $ alrededor del eje $z$.

${}$

Ejemplo 2

Dada $f (x , y) = x^5 y + x y^5 + xy $

Analicemos cuando las derivadas parciales valen cero.

$\dfrac{\partial f}{\partial x} = 5 x^4 y + y^5 + y = 0 \Rightarrow (5 x^4 + y^4 + 1) y = 0 \Rightarrow y = 0$

$\dfrac{\partial f}{\partial y} = 5 y^4 x + x^5 + x = 0 \Rightarrow (5 y^4 + x^4 + 1) x = 0 \Rightarrow x = 0$

Por lo que el punto $(0, 0)$ es el único punto crítico.

El valor crítico correspondiente es

$f (0, 0) = 0$

Segundas derivadas parciales

$\dfrac{\partial^2 f}{\partial x^2} = 20 x^3 y $

Luego $f_{xx} (0, 0) = 0$

Análogamente

$\dfrac{\partial^2 f}{\partial y^2} = 20 y^3 x $

Luego $f_{yy} (0, 0) = 0$

Además

$\dfrac{\partial^2 f}{\partial y \, \partial x} = 5 x^4 + 5 y^4 + 1 $

Luego $f_{xy} (0, 0) = 1$

En el $(0, 0)$ la matriz queda de la siguiente manera

$H = \begin{pmatrix} f_{xx} & f_{xy} \\ \\ f_{xy} & f_{yy} \end{pmatrix} = \begin{pmatrix} 0 & 1 \\ \\ 1 & 0 \end{pmatrix}$

Luego el $det (H) < 0$ por lo tanto el punto $(0, 0)$ es punto silla.

Además, polinomio de Taylor de 2° grado de $f$ alrededor del punto $( 0, 0)$ es

$$p ( x, y) = x y $$

En cierto sentido, así tenía que ser ya que $f (x, y) = x^5 y + x y^5 + x y \approx x y = p (x, y) $ cerca del $( 0, 0)$.

${}$

Ejemplo 3

La gráfica de la función $f (x , y) = \dfrac{1}{xy} $ es una superficie $\mathcal{S}$ que NO contiene al origen.

$$\mathcal{S} = \Big\{ (x, y, z) \in \mathbb{R}^3 \big| z = \dfrac{1}{xy} , x \neq 0, y \neq 0 \Big\}$$

Queremos saber cuáles son los puntos de $\mathcal{S}$ que son más cercanos al origen, lo que es equivalente a minimizar la distancia al origen.

Entonces

$\Bigg\| \Bigg( x, y, \dfrac{1}{xy} \Bigg) \, – \, (0, 0, 0) \Bigg\| = \sqrt{x^2 + y^2 + \dfrac{1}{x^2 y^2} \, } $

pero minimizar $ f (x, y ) \geq 0 $ es equivalente a minimizar $f^2 (x, y) $

Entonces tenemos que minimizar $ F (x, y) = x^2 + y^2 + \dfrac{1}{x^2 y^2}$.

Precaución: $x \neq 0 $ y también $ y \neq 0 $; por lo que el dominio no es todo $\mathbb{R}^2.$

Analicemos cuando las derivadas parciales valen cero

$\dfrac{\partial f}{\partial x} = 2x + (-2) x^{-3} y^{-2} = 2x \, – \, \dfrac{2}{x^3 y^2} \Rightarrow 2x = \dfrac{2}{x^3 y^2} \Rightarrow x^4 y^2 = 1 $… (1)

$\dfrac{\partial f}{\partial y} = 2y + (-2) x^{-2} y^{-3} = 2y \, – \, \dfrac{2}{x^2 y^3} \Rightarrow 2y = \dfrac{2}{x^2 y^3} \Rightarrow x^2 y^4 = 1$

Por lo que $ x^2 y^4 = x^4 y^2 $ dado que $xy \neq 0 \; \Rightarrow \dfrac{x^2 y^4}{x^2 y^2} = \dfrac{x^4 y^2}{x^2 y^2} \Rightarrow x^2 = y^2 $… (2)

Sustituyendo según (2) en la expresión (1) tenemos que

$ x^4 y^2 = 1 \Rightarrow x^6 = 1 \Rightarrow x = \pm 1$ y por lo tanto $y = \pm 1$.

Por lo que se tienen cuatro puntos críticos, los puntos que corresponden a $( \pm 1, \pm 1).$

Observación

$F (\pm x, \pm y) = F (x, y) $ por lo que la gráfica de $F$ es simétrica respecto al plano $x = 0$ y respecto al plano $y = 0.$ Es decir, basta examinar lo que sucede en la región $x \geq 0 $ y $ y \geq 0.$

Los valores críticos correspondientes son

$F (\pm 1, \pm 1) = (\pm 1)^2 + (\pm 1)^2 + \dfrac{1}{(\pm 1)^2(\pm 1)^2} = 1 + 1 + 1 = 3$

Consideremos uno de los puntos críticos, por ejemplo el punto $( 1, 1)$. ¿Cuál es el polinomio de Taylor que aproxima a $F (x, y)$ cerca del punto $( 1, 1)$?

Calculamos las primeras derivadas parciales

$\dfrac{\partial F}{\partial x} = 2x \, – \, \dfrac{2}{x^3 y^2}$

$\dfrac{\partial F}{\partial y} = 2y \, – \, \dfrac{2}{x^2 y^3}$

Segundas derivadas parciales

$\dfrac{\partial^2 F}{\partial x^2} = 2 + \dfrac{6}{x^4 y^2} $

Luego $f_{xx} (1, 1) = 8$

Análogamente

$\dfrac{\partial^2 F}{\partial y^2} = 2 + \dfrac{6}{x^2 y^4} $

Luego $f_{yy} (1, 1) = 8$

Además

$\dfrac{\partial^2 F}{\partial y \, \partial x} = \dfrac{4}{x^3 y^3} $

Luego $f_{xy} (1, 1) = 4$

Entonces la matriz $H$ queda de la siguiente manera:

$$\begin{vmatrix} f_{xx} (1, 1) & f_{xy} (1, 1) \\ \\ f_{xy} (1, 1) & f_{yy} (1, 1) \end{vmatrix} = \begin{pmatrix} 8 & 4 \\ \\ 4 & 8 \end{pmatrix} = 4 \begin{pmatrix} 2 & 1 \\ \\ 1 & 2\end{pmatrix}$$

El determinante $det(H) = 3 > 0$

Y la traza es $4$

Por lo tanto, $F$ alcanza un valor mínimo en $ ( 1, 1).$

Entonces el polinomio de Taylor de 2° grado de $F$ alrededor del punto $( 1, 1)$ es

$$p ( x, y) = F ( 1, 1) + \dfrac{\partial F}{\partial x} ( 1, 1) (x \, – \, 1) + \dfrac{\partial F}{\partial y} ( 1, 1) (y \, – \, 1) + \dfrac{1}{2} \Big( \dfrac{\partial^2 F}{\partial x^2} ( 1, 1) (x \, – \, 1)^2 + 2 \dfrac{\partial^2 F}{\partial x \, \partial y} ( 1, 1) (x \, – \, 1) (y \, – \, 1) + \dfrac{\partial^2 F}{\partial y^2} ( 1, 1) (y \, – \, 1) \Big)$$

$$p ( x, y) = 3 + \dfrac{1}{2}\Big( 8 (x \, – \, 1)^2 + 8 (x \, – \, 1) (y \, – \, 1) + 8 (y \, – \, 1)^2 \Big)$$

$$p ( x, y) = 3 + 4 \Big( (x \, – \, 1)^2 + (x \, – \, 1) (y \, – \, 1) + (y \, – \, 1)^2 \Big)$$

Luego, uno de los puntos de $\mathcal{S}$ más cercanos al $(0, 0)$ es el punto $(1, 1, 1)$, así como también los puntos $(1, \, – \,1, \, – \,1)$, $(\, – \, 1, \, – \, 1, 1)$ , y $(\, – \, 1, 1, \, – \, 1)$.

La distancia desde cada uno de estos puntos al origen es $\sqrt{3}.$

60. Material en revisión: viernes 18 de octubre

Por Mariana Perez

Consideremos una curva parametrizada $$\alpha : I \subseteq \mathbb{R} \rightarrow \mathbb{R}^2$$ $$\alpha (t) = \big( x (t), y (t) \big)$$

Supongamos que $ t = 0 $

$ x ( 0 ) = 0 $

$ y ( 0 ) = 0 $

$\alpha ( 0 ) = ( 0 , 0 )$

Supongamos además que $x $, $ y $ son funciones derivables de $ t $ y que $\big( x’ (t), y’ (t) \big) \neq ( 0, 0) $ vector velocidad.

Llamemos $\vec{v} = \big( x’ (0), y’ (0) \big)$

Consideremos una reparametrización de $\alpha$ $$\beta (s) = \alpha ( h (s) )$$

con $\beta = \alpha \, \circ \, h$, para algún $ h : J \subseteq \mathbb{R} \rightarrow \mathbb{R}$.

Sin pérdida de generalidad, supongamos también $ h (0) = 0$

${\beta \,}’ (s) = \alpha’ ( h (s) ) \cdot h’ ( s )$

${\beta \,}’ (s) = \big( x’ (h (s)), y’ (h (s)) \big) \cdot h’ ( s )$

$\vec{w} = {\beta \,}’ (0) = \big( x’ (0), y’ (0) \big) \cdot h’ (0) = h’ (0) \cdot \vec{v}$

Pedimos que $ h’ (0) \neq 0$

$ \vec{w} $ es el vector velocidad usando $\beta$ como parametrización.

Los dos vectores velocidad $ \vec{v}$ y $\vec{w} $ son colineales. Además, si

$ h’ (0) > 0 $ tienen el mismo sentido, y si

$h’ (0) < 0 $ tienen sentidos contrarios.

${}$

Supongamos que $ \gamma (t) = \big( x (t), y (t) \big)$ es una curva parametrizada, y que $ x (t)$ , $y (t) $ son de clase $\mathcal{C}^2$.

Además

$\gamma’ (0) \neq \vec{0}$

${\gamma}^{\prime \prime} (0) \neq \vec{0}$

¿Cómo saber cuál es la circunferencia osculatriz en el punto $P$?

Si $\gamma $ estuviera parametrizada por longitud de arco, entonces el centro de la circunferencia osculatriz es $$ \gamma (0) + \dfrac{1}{\mathcal{K} (0)} \vec{n} (0)$$

donde $\vec{n} (0) $ es el vector normal unitario.

Y el radio de la circunferencia osculatriz es $\dfrac{1}{\mathcal{K} (0)}$

Sabiendo la curvatura y el vector normal tenemos toda la información necesaria. (no necesariamente tiene que ser en el punto CERO, puede ser un punto $t_0$ con $\gamma’ (t_0) $ y ${\gamma}^{\prime \prime} (t_0)$ distintas de CERO)

(*) Tenemos una fórmula para calcular $\mathcal{K} (t_0)$

(*) a. En el plano, basta conocer el vector tangente unitario $T (t_0) = \dfrac{ \gamma’ (t_0)}{ \big\| \gamma’ (t_0) \big\|}$. Solo hay dos opciones para $(u, v)$ puede ser $ (- \, v, u ) $ o también $ (v, \, – \, u) $

b. En el espacio

$$ \overrightarrow{N} (t) = \dfrac{\dfrac{d}{dt} \overrightarrow{T} (t) }{\Bigg\| \dfrac{d}{dt} \overrightarrow{T} (t) \Bigg\|}$$

59. Material en revisión: (miércoles 16 de octubre)

Por Mariana Perez

(*) Sobre las derivadas de $g (t) = f (x_0 + th, y_0 + tk)$

Proposición:

$$ g^{(n)} (t) = \sum\limits_{i=0}^{n} \begin{pmatrix} n \\ i \end{pmatrix} \dfrac{\partial^n f}{\partial x^i \; \partial y^{n-i}} (x_0 + th, y_0 + tk) h^i \; k^{n-i}$$

Demostración (por inducción)

Base inductiva:

Para $n = 1, 2 $ lo trabajamos en la entrada anterior. https://blog.nekomath.com/?p=101046&preview=true

Paso inductivo:

Supongamos la proposición válida para $n.$

$\Big[$ por demostrar: que es válida para $ n + 1 \Big]$

Entonces

$\begin{align*} g^{(n+1)} (t) &= \dfrac{d}{dt} \sum\limits_{i=0}^{n} \begin{pmatrix} n \\ i \end{pmatrix} \dfrac{\partial^n f}{\partial x^{i} \; \partial y^{n-i}} (x_0 + th, y_0 + tk) h^{i} \; k^{n-i} \\ &= \sum\limits_{i=0}^{n} \begin{pmatrix} n \\ i \end{pmatrix} h^{i} \; k^{n-i} \dfrac{d}{dt} \dfrac{\partial^n f}{\partial x^{i} \; \partial y^{n-i}} (x_0 + th, y_0 + tk) \end{align*}$

Podemos ver a la $\dfrac{\partial^n f}{\partial x^{i} \; \partial y^{n-i}} (x, y) = G (x, y)$

Entonces, derivando $G (x_0 + th, y_0 + tk) = G ( \alpha ( t)) $ respecto a $ t $ se tiene que

$\dfrac{d}{dt} \dfrac{\partial^n f}{\partial x^{i} \; \partial y^{n-i}} (x_0 + th, y_0 + tk) = \dfrac{\partial^{n+1} f}{\partial x^{i+1} \; \partial y^{n-i}} (x_0 + th, y_0 + tk) h \, + \, \dfrac{\partial^{n+1} f}{\partial x^{i} \; \partial y^{n-i+1}} (x_0 + th, y_0 + tk) k$

Entonces

$\begin{align*} g^{(n+1)} (t) &= \sum\limits_{i=0}^{n} \begin{pmatrix} n \\ i \end{pmatrix} h^{i} \; k^{n-i} \Bigg( \dfrac{\partial^{n+1} f}{\partial x^{i+1} \; \partial y^{n-i}} (x_0 + th, y_0 + tk) h \, + \, \dfrac{\partial^{n+1} f}{\partial x^{i} \; \partial y^{n-i+1}} (x_0 + th, y_0 + tk) k \Bigg) \end{align*}$…(1)

$\Big[$ por demostrar: $g^{(n + 1)} (t) = \sum\limits_{i=0}^{n+1} \begin{pmatrix} n+1 \\ i \end{pmatrix} h^{i} \; k^{n+1-i} \dfrac{\partial^{n+1} f}{\partial x^{i} \; \partial y^{n+1-i}} (x_0 + th, y_0 + tk) \Big]$

De $(1)$ tenemos que

$g^{(n+1)} (t) = \sum\limits_{i=0}^{n} \begin{pmatrix} n \\ i \end{pmatrix} h^{i} \; k^{n-i} \dfrac{\partial^{n+1} f}{\partial x^{i+1} \; \partial y^{n-i}} \, + \, \sum\limits_{i=0}^{n} \begin{pmatrix} n \\ i \end{pmatrix} h^{i} \; k^{n+1-i} \dfrac{\partial^{n+1} f}{\partial x^{i} \; \partial y^{n+1-i}} $

haciendo $ j = i + 1$

$g^{(n+1)} (t) = \sum\limits_{j=1}^{n+1} \begin{pmatrix} n \\ j-1 \end{pmatrix} h^{i} \; k^{n+1-i} \dfrac{\partial^{n+1} f}{\partial x^{j} \; \partial y^{n+1-j}} \, + \, \sum\limits_{i=0}^{n} \begin{pmatrix} n \\ i \end{pmatrix} h^{i} \; k^{n+1-i} \dfrac{\partial^{n+1} f}{\partial x^{i} \; \partial y^{n+1-i}} $

con $ i = 0$, $ j = n + 1$

$ g^{(n+1)} (t) = \begin{pmatrix} n \\ 0 \end{pmatrix} h^{0} \; k^{n+1} \dfrac{\partial^{n+1} f}{ \partial y^{n+1}} \, + \, \begin{pmatrix} n \\ n \end{pmatrix} h^{n+1} \; k^{0} \dfrac{\partial^{n+1} f}{\partial x^{n+1} } \, + \, \sum\limits_{i=1}^{n} \Bigg[ \begin{pmatrix} n \\ i-1 \end{pmatrix} , + \, \begin{pmatrix} n \\ i \end{pmatrix} \Bigg] h^{i} \; k^{n+1-i} \dfrac{\partial^{n+1} f}{\partial x^{i} \; \partial y^{n+1-i}} $

entonces

$ g^{(n+1)} (t) = \begin{pmatrix} n \\ 0 \end{pmatrix} h^{0} \; k^{n+1} \dfrac{\partial^{n+1} f}{ \partial y^{n+1}} \, + \, \begin{pmatrix} n \\ n \end{pmatrix} h^{n+1} \; k^{0} \dfrac{\partial^{n+1} f}{\partial x^{n+1} } \, + \, \sum\limits_{i=1}^{n} \begin{pmatrix} n + 1 \\ i \end{pmatrix} h^{i} \; k^{n+1-i} \dfrac{\partial^{n+1} f}{\partial x^{i} \; \partial y^{n+1-i}} $

${}$

(*) Aplicamos el teorema de Taylor a la función $ g (t) = f (x_0 + th, y_0 + tk) $ en $[ 0, 1]$

Teorema de Taylor en una variable

$f (x) = f (a) + f’ (a) (x\, – \, a) + \dfrac{ {f’}’ (a)}{2} (x\, – \, a)^2 \, + \, R_2 (x)$

Fórmulas para $R_2 (x)$

(1) Lagrange: $R_2 (x) = \dfrac{ f^{(3)} (t) }{3!} (x\, – \, a)^3$ para algún $ t \in ( a, x)$.

(2) Integral: $R_2 (x) = \int\limits_{a}^{x} \dfrac{ f^{(3)} (t) }{2!} (x\, – \, a)^2 dt$

Si $[a, x] = [0, 1]$ entonces $ x \, – \, a = 1$, por lo que

$ g (1) = g (0) \, + \, g’ (0) \, + \, \dfrac{ {g’}’ (0) }{2!} \, + \, R_2 (1)$

Si $(x_0, y_0)$ es un punto crítico, entonces

$f (x_0 + h, y_0 + k) = f (x_0 , y_0) \, + \, \cancel{\dfrac{\partial f}{\partial x} (x_0, y_0) h } \, + \, \cancel{ \dfrac{\partial f}{\partial y} (x_0, y_0) k} \, + \, \dfrac{1}{2} \Bigg[ \textcolor{Magenta}{\dfrac{\partial^2 f}{\partial x^2} (x_0, y_0) h^2 \, + \, 2 \dfrac{\partial^2 f}{\partial x \; \partial y} (x_0, y_0) h\, k \, + \, \dfrac{\partial^2 f}{\partial y^2} (x_0, y_0) k^2} \Bigg] \, + \, R_2 (1) $

La expresión resaltada de color se conoce como $\textcolor{Magenta}{Forma \; \; cuadr\acute{a}tica}.$

Calculamos el error con cualquiera de las dos fórmulas vistas, de modo que:

(1) $\dfrac{g^{(3)} (t)}{3!} = \frac{1}{3!} \Bigg( \dfrac{ \partial f^{(3)}}{\partial x^3} h^3 \, + \, 3 \dfrac{\partial^3 f}{\partial x^2 \; \partial y} h^2 \; k \, + \, 3 \dfrac{\partial^3 f}{\partial x \; \partial y^2} h \; k^2 \, + \, \dfrac{ \partial f^{(3)}}{\partial y^3} k^3 \Bigg)$ en $ t \in (0, 1) $

(2) $\int\limits_{0}^{1} \dfrac{g^{(3)} (t)}{2} (1 \, – \, t)^2 dt$

${}$

(*) Un acercamiento a las formas cuadráticas.

Sea $H : \mathbb{R}^2 \rightarrow \mathbb{R}$, donde

$H (x, y) = \begin{pmatrix} x & y \end{pmatrix} \begin{pmatrix} a & b \\ b & c \end{pmatrix} \begin{pmatrix} x \\ y \end{pmatrix}$

$H (x, y) = a x^2 + 2bxy + c y^2$

Definición: $H $ es definida positiva ( o positivamente) si

$\begin{cases} H (x, y) > 0 \; \; \; \forall \; (x, y) \neq (0, 0) \\ {} \\ H (x, y) = 0 \; \; \; si \; (x, y) = (0, 0) \end{cases}$

Análogamente,

$H $ es definida negativa ( o negativamente) si

$ H (x, y) < 0 \; \; \; \forall \; (x, y) \neq (0, 0)$

Además

$H$ es semi definida positiva, si $H (x, y) \geq 0 \forall \; (x, y)$

$H$ es semi definida negativa, si $H (x, y) \leq 0 \forall \; (x, y)$

Las formas cuadráticas más fáciles de estudiar son las que tienen asociada una matriz diagonal $$\begin{pmatrix} \lambda_1 & 0 \\ \\ 0 & \lambda_2 \end{pmatrix}$$

Entonces

$$\begin{equation*} H (x, y) = \begin{pmatrix} x & y \end{pmatrix} \begin{pmatrix} \lambda_1 & 0 \\ \\ 0 & \lambda_2 \end{pmatrix} \begin{pmatrix} x \\ y \end{pmatrix} \end{equation*}$$

$$\begin{equation*} H (x, y) = \lambda_1 x^2 + \lambda_2 y^2 \end{equation*}$$

$H$ es definida positiva si $\begin{cases} \lambda_1 > 0 \\ \lambda_2 > 0 \end{cases}$

$H$ es definida negativa si $\begin{cases} \lambda_1 < 0 \\ \lambda_2 < 0 \end{cases}$

$H$ es semi definida positiva si $\begin{cases} \lambda_1 \geq 0 \\ \lambda_2 \geq 0 \end{cases}$

$H$ es semi definida negativa si $\begin{cases} \lambda_1 \leq 0 \\ \lambda_2 \leq 0 \end{cases}$

$H$ tiene un punto silla en el origen si $ \lambda_1 < 0 < \lambda_2$

${}$

Veamos un ejemplo:

$H (x, y) = 2xy$

Consideremos composiciones $H \, \circ \, T$ , con $T : \mathbb{R}^2 \rightarrow \mathbb{R}^2$ lineal invertible.

$$ \begin{pmatrix} x’ \\ y’ \end{pmatrix} \rightarrow \begin{pmatrix} x \\ y \end{pmatrix}$$

$T (x’ , y’ ) = ( x , y )$

$\begin{pmatrix} a & b \\ b & c \end{pmatrix} \begin{pmatrix} x’ \\ y’ \end{pmatrix} = \begin{pmatrix} x \\ y \end{pmatrix}$

$\begin{cases} x = ax’ + b y’ \\ y = c x’ + d y’ \end{cases}$

$H \big( T ( x’ , y’ ) \big) = H ( x , y )$

$\begin{align*}H \big( T ( x’ , y’ ) \big) &= 2 (ax’ + b y’ ) ( c x’ + d y’) \\ &= 2 ac ( x’ )^2 + 2 ( ad + bc ) x’ y’ + 2 bd ( y’ )^2 \end{align*}$

Buscamos elegir $T$ tal que $ ad + bc = 0$

$ H \, \circ \, T$ tiene asociada la matriz

$\begin{pmatrix} 2 ac & ad + bc \\ ad + bc & 2 bd \end{pmatrix}$…(1)

Ahora bien, si $\vec{v} = \begin{pmatrix} x \\ y \end{pmatrix}$ y $\vec{w} = \begin{pmatrix} x’ \\ y’ \end{pmatrix}$

$A \vec{w} = \vec{v}$

$H ( \vec{v} ) = \vec{v}^t M \vec{v}$, donde $M = \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix}$

Entonces

$\begin{align*} H ( \vec{v} ) &= \vec{v}^t M A \vec{w} \\ &= \big( A \vec{w} \big)^t M A \vec{w} \\ &= \vec{w}^t A^t M A \vec{w} \\ &= A^t M A \end{align*}$

Comprobamos que (1) $= A^t M A$

$\begin{align*} A^t M A &= \begin{pmatrix} a & c \\ b & d \end{pmatrix} \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix} \begin{pmatrix} a & b \\ c & d \end{pmatrix} \\ \\ &= \begin{pmatrix} a & c \\ b & d \end{pmatrix} \begin{pmatrix} c & d \\ a & b \end{pmatrix} \\ \\ A^t M A &= \begin{pmatrix} 2 ac & ad + bc \\ ad + bc & 2 bd \end{pmatrix}\end{align*}$

Buscamos $T$ tal que $ A^t M A $ sea diagonal.

Si $A = \begin{pmatrix} \cos \theta & – \, \sin \theta \\ \\ \sin \theta & \cos \theta \end{pmatrix}$ matriz de rotación de ejes.

$ A^t = \begin{pmatrix} \cos \theta &\sin \theta \\ \\ – \, \sin \theta & \cos \theta \end{pmatrix} = \begin{pmatrix} \cos ( – \, \theta) & \, – \, \sin ( – \, \theta) \\ \\ \sin ( – \, \theta) & \cos ( – \, \theta) \end{pmatrix} = A^{-1}$

Luego $ A^t M A = A^{-1} M A$ … (2)

Sabemos entonces que $A$ es inyectiva, invertible y su determinante $det(A) = 1 \neq 0$

Si $\vec{v_1} = \begin{pmatrix} \cos \theta \\ \\ \sin \theta \end{pmatrix} $

$\vec{v_2} = \begin{pmatrix} – \, \sin \theta \\ \\ \cos \theta \end{pmatrix} $

Entonces $A \vec{e_1} = \vec{v_1} $ y también $A \vec{e_2} = \vec{v_2} $, si además $\vec{v_1}$, $\vec{v_2}$ son eigenvectores ( o vectores propios) de $M$, entonces

$M \vec{v_1} = \lambda_1 \vec{v_1}$

$M \vec{v_2} = \lambda_2 \vec{v_2}$

$ A^{-1} M A \vec{e_1} = A^{-1} M \vec{v_1} = A^{-1} \lambda_1 \vec{v_1} = \lambda_1 A^{-1} \vec{v_1} = \lambda_1 \vec{e_1} $

Análogamente

$ A^{-1} M A \vec{e_2} = A^{-1} M \vec{v_2} = A^{-1} \lambda_2 \vec{v_2} = \lambda_2 A^{-1} \vec{v_2} = \lambda_2 \vec{e_2} $

Luego (2) es igual a $\begin{pmatrix} \lambda_1 & 0 \\ \\ 0 & \lambda_2 \end{pmatrix}$

En nuestro ejemplo:

$\vec{v_1} = \begin{pmatrix} \dfrac{1}{\sqrt{2}} \\ \\ \dfrac{1}{\sqrt{2}} \end{pmatrix}$

$\vec{v_2} = \begin{pmatrix} \dfrac{-1}{\sqrt{2}} \\ \\ \dfrac{1}{\sqrt{2}} \end{pmatrix}$

Además $ \lambda_1 = 1 $ y $\lambda_2 = \, – \, 1$

Luego $( x’ )^2 \, – \, ( y’ )^2 = 2xy = H ( x , y ) $

58. Material en revisión: Derivadas direccionales iteradas y teorema del valor medio para la derivada. (martes 15 de octubre)

Por Mariana Perez

Sea $ f : U \subseteq \mathbb{R}^2 \rightarrow \mathbb{R}$ , diferneciable en un abierto, convexo $ U. $

Sea $ (x_0, y_0) \in U$ y $ (h, k) $ cercano a $ (x_0, y_0) $, tal que $(x_0 + h , y_0 + k) \in U.$

Como $ U $ es convexo, entonces $$ (1 \, – \,t) (x_0, y_0) + t ( x_0 + h , y_0 + k) = (x_0 + t h , y_0 + t k) \in U$$

Podemos parametrizar el segmento de recta $ \alpha (t) = (x_0 + t h, y_0 + t k) $ y hacer la composición $ f \, \circ \, \alpha : [ 0, 1] \rightarrow \mathbb{R}$

Le podemos aplicar:

(*) el teorema del valor medio para funciones de $\mathbb{R} \rightarrow \mathbb{R}$, y también

(*) la regla de la cadena.

${}$

Teorema del valor medio para derivadas

Si $ g $ es continua en el cerrado $[0, 1]$ y derivable en el abierto $(0, 1)$, entonces existe $\theta \in (0, 1) $ tal que

$$g’ (\theta) = \dfrac{g (1) \, – \, g (0) }{1 \, – \, 0} = g (1) \, – \, g (0) $$

donde $ g = f \, \circ \, \alpha$

$ ( f \, \circ \, \alpha ) (\theta) = \nabla f ( \alpha ( \theta )) \cdot {\alpha}’ (\theta) = \dfrac{\partial f}{\partial x} (x_0 + \theta h , y_0 + \theta k) h \, + \, \dfrac{\partial f}{\partial y} (x_0 + \theta h , y_0 + \theta k) k$

Sabemos que

$ f (x_0 + h , y_0 + k) = f (x_0 , y_0) + \nabla f ( \alpha ( \theta )) \cdot {\alpha}’ (\theta) $

$ f (x_0 + h , y_0 + k) \, – \, f (x_0 , y_0) = \nabla f ( \alpha ( \theta )) \cdot {\alpha}’ (\theta) $

Si $ f $ fuera de clase $\mathcal{C}^2$ podríamos decir más, en particular $ g = f \, \circ \, \alpha \in \mathcal{C}^2$

$g (t) = f ( x_0 + t h, y_0 + t k)$

$g’ (t) = \dfrac{\partial f}{\partial x} ( x_0 + t h, y_0 + t k) h \, + \, \dfrac{\partial f}{\partial y} ( x_0 + t h, y_0 + t k) k$

Queremos conocer calcular $ {g}^{\prime \prime} (t)$.

Examinemos $\dfrac{\partial f}{\partial x} ( x_0 + t h, y_0 + t k) h = G ( x_0 + t h, y_0 + t k)$

Sea $G (x, y) = \dfrac{\partial f}{\partial x} (x, y)$, entonces

$ \dfrac{d}{dx} G ( x_0 + t h, y_0 + t k) = \dfrac{\partial G}{\partial x} ( x_0 + t h, y_0 + t k) h \, + \, \dfrac{\partial G}{\partial y} k$

Entonces

$$ {g}^{\prime \prime} = \Bigg( \dfrac{\partial^2 f}{\partial x^2} ( x_0 + t h, y_0 + t k) h \, + \, \dfrac{\partial f}{\partial y \partial x} ( x_0 + t h, y_0 + t k) k \Bigg) h \, + \, \Bigg( \dfrac{\partial f}{\partial x \partial y} ( x_0 + t h, y_0 + t k) h \, + \, \dfrac{\partial^2 f}{\partial y^2} ( x_0 + t h, y_0 + t k) k \Bigg) k$$

$${g}^{\prime \prime} = \dfrac{\partial^2 f}{\partial x^2} ( x_0 + t h, y_0 + t k) h^2 \, + \, 2 \dfrac{\partial^2 f}{\partial y \partial x} ( x_0 + t h, y_0 + t k) h k \, + \, \dfrac{\partial^2 f}{\partial y^2} ( x_0 + t h, y_0 + t k) k^2 $$

${g}^{\prime \prime} $ es la segunda derivada de $ ( f \, o \, \alpha) ( t ).$

La tercera derivada de $ f ( \alpha ( t ) )$ es $$ \dfrac{\partial^3 f}{\partial x^3} h^3 \, + \, 3 \dfrac{\partial^3 f}{\partial x^2 \partial y} h^2 k \, + \, 3 \dfrac{\partial^3 f}{\partial x \partial y^2} h k^2 \, + \, \dfrac{\partial^3 f}{\partial y^3} k^3 $$ valuada en $( x_0 + t h, y_0 + t k)$

${}$

Teorema de Taylor para funciones de $\mathbb{R} \rightarrow \mathbb{R}$ aplicado a $ g = f \, \circ \, \alpha$

$g ( t ) = g ( 0 ) + g’ ( 0 ) t + \frac{1}{2} {g}^{\prime \prime} ( 0 ) t^2 + E ( t )$ , donde $ E ( t )$ es el error.

Una fórmula para este error es $ \dfrac{{g}^{\prime \prime \prime} ( \xi ) }{ 3! } $ para alguna $ \xi \in ( 0, t)$

Entonces

$ g ( t ) = f ( x_0 + t h, y_0 + t k)$

$ g ( 0 ) = f ( x_0 , y_0 )$

$ g’ ( 0 ) = \dfrac{\partial f}{\partial x} ( x_0 , y_0 ) h + \dfrac{\partial f}{\partial y} ( x_0 , y_0 ) k$

$ {g}^{\prime \prime} ( 0 ) = \dfrac{\partial^2 f}{\partial x} ( x_0 , y_0 ) h^2 + 2 \dfrac{\partial^2 f}{\partial y \partial } h k + \dfrac{\partial^2 f}{\partial y^2} ( x_0 , y_0 ) k^2 $

$ E (t) = \dfrac{{g}^{\prime \prime \prime} ( \xi)}{3!} = \dfrac{\dfrac{\partial^3 f}{\partial x^3} h^3 \, + \, 3 \dfrac{\partial^3 f}{\partial x^2 \partial y} h^2 k \, + \, 3 \dfrac{\partial^3 f}{\partial x \partial y^2} h k^2 \, + \, \dfrac{\partial^3 f}{\partial y^3}}{3!}$

Si $f $ es de clase $\mathcal{C}^3$ ya tenemos un polinomio de 2° grado que aproxima bien a $f$ localmente.

Si el punto es un $\textit{ punto crítico }$ entonces, el polinomio de 2° grado es, esencialmente, un polinomio homogéneo.

$$ \mathbb{R}^2 \rightarrow \mathbb{R} \; \; \; ax^2 + 2bxy + cy^2$$

$ p (x, y) = ax^2 + 2bxy + cy^2 = (x , y ) \begin{equation*} \begin{pmatrix} a & b \\ b & c \end{pmatrix} \end{equation*}$

$ p ( 0 ) = 0$

$ \nabla p ( 0, 0) = \Big( \dfrac{\partial p }{\partial x} , \dfrac{\partial p }{\partial y} \Big) = ( 2ax + 2 by, 2 bx + 2 cy ) \Biggm|_{(0, 0)} = (0, 0)$

El plano tangente a $ z = p (x, y )$ es horizontal.

Observación:

La matriz $ A = \begin{equation*} \begin{pmatrix} a & b \\ b & c \end{pmatrix} \end{equation*}$ es simétrica.

También sabemos que si la matriz es diagonalizable, entonces existe una base ortonormal de vectores propios $\vec{v_1}, \, \vec{v_2}$ tales que

$\| \vec{v_1} \| = \| \vec{v_2} \| = 1$ y también $ \vec{v_1} \cdot \vec{v_2} = 0$

Y existe $\theta$ tal que

$ \vec{v_1} = (\cos \theta , \sin \theta ) $

$ \vec{v_2} = (\, – \ \sin \theta, \cos \theta) $

Además, existen $\lambda_1 , \, \lambda_2$ tales que $$ A \vec{v_1} = \lambda_1 \vec{v_1} $$ $$ A \vec{v_2} = \lambda_2 \vec{v_2} $$

$D = \begin{equation*} \begin{pmatrix} \lambda_1 & 0 \\ 0 & \lambda_2 \end{pmatrix} \end{equation*}$ es la matriz diagonal.

Sin pérdida de generalidad en al caso cuando

$\begin{equation*} \begin{pmatrix} a & b \\ b & c \end{pmatrix} = \begin{pmatrix} \lambda_1 & 0 \\ 0 & \lambda_2 \end{pmatrix} \end{equation*}$

$ p (x, y) = \lambda_1 x^2 + \lambda_2 y^2$

Analicemos los tres casos posibles según los valores de $\lambda$

CASO 1: $ \lambda_1 , \lambda_2 > 0$ entonces, $p$ alcanza un valor mínimo.

https://www.geogebra.org/classic/drtwmcwt

Ejemplo: $p (x, y) = x^2 + y^2 $

CASO 2: $ \lambda_1 , \lambda_2 < 0$ entonces, $p$ alcanza un valor máximo.

https://www.geogebra.org/classic/zkkwuga4

Ejemplo: $p (x, y) = \, – \, x^2 \, – \, y^2 $

CASO 3: $ \lambda_1 > 0 , \lambda_2 < 0$ entonces, $p$ tiene un punto silla.

https://www.geogebra.org/classic/yczbtnvb

Ejemplo: $p (x, y) = x^2 \, – \, y^2 $

Si alguno de los $\lambda$ es CERO, no se puede concluir nada acerca del punto.

${}$

$\underline{\textit{Veamos un ejemplo}}:$

$p (x, y) = 2xy$

Como $ p (x, y) = ax^2 + 2bxy + cy^2$ entonces, para nuestro ejemplo tenemos que:

$ a = 0$ , $ b = 1$ , y $ c = 0$.

Entonces la matriz $A = \begin{equation*} \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix} \end{equation*}$

Entonces $\begin{equation*} \begin{vmatrix} – \, \lambda & 1 \\ 1 & – \, \lambda \end{vmatrix} \end{equation*} = 0$

$ \lambda^2 \, – \, 1 = 0 $

$ \lambda^2 = 1$ y por tanto $ \lambda = \pm 1$

$\underline{\textit{Vectores propios}}:$

$$\begin{equation*} \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix} \begin{pmatrix} x \\ y \end{pmatrix} = 1 \begin{pmatrix} x \\ y \end{pmatrix} \end{equation*}$$

Entonces

$$\begin{equation*} \begin{pmatrix} y \\ x \end{pmatrix} = \begin{pmatrix} x \\ y \end{pmatrix} \end{equation*}$$

Luego $ \vec{v_1} = \dfrac{( 1, 1)}{ \| \vec{v_1} \|} = \begin{pmatrix} \dfrac{1}{\sqrt{2}} \\ {} \\ \dfrac{1}{\sqrt{2}} \end{pmatrix} = \begin{pmatrix} \cos \dfrac{\pi}{4} \\ {} \\ \sin \dfrac{\pi}{4} \end{pmatrix}$

Análogamente,

$$\begin{equation*} \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix} \begin{pmatrix} x \\ y \end{pmatrix} = \, – \, 1 \begin{pmatrix} x \\ y \end{pmatrix} \end{equation*}$$

Entonces

$$\begin{equation*} \begin{pmatrix} y \\ x \end{pmatrix} = \begin{pmatrix} \, – \, x \\ \, – \, y \end{pmatrix} \end{equation*}$$

Luego $ \vec{v_2} = \begin{pmatrix} \dfrac{ \, – \, 1}{\sqrt{2}} \\ {} \\ \dfrac{ \, – \, 1}{\sqrt{2}} \end{pmatrix} = \begin{pmatrix} \, – \, \sin \dfrac{\pi}{4} \\ {} \\ \cos \dfrac{\pi}{4} \end{pmatrix}$

${}$

Corte con el plano vertical $ x = y$ son las parábolas $p (x, y) = 2x^2$

Corte con el plano vertical $ y = \, – \, x$ son las parábolas $p (x, y) = \, – \, 2x^2$

En el siguiente enlace puedes observar la gráfica de la superficie $p (x,y)$ y sus cortes con los planos verticales.

https://www.geogebra.org/classic/rjhgjzeq