תרגיל 1
תהי g ( x , y ) g(x,y) g ( x , y ) פונקציה בעלת נגזרות חלקיות רציפות בתחום פתוח D D D .
תהי ( x 0 , y 0 ) ∈ D (x_{0},y_{0})\in D ( x 0 , y 0 ) ∈ D נקודת מקסימום מקומי של g g g , כלומר g ( x 0 , y 0 ) ≥ g ( x , y ) g(x_{0},y_{0})\geq g(x,y) g ( x 0 , y 0 ) ≥ g ( x , y ) לכל ( x , y ) (x,y) ( x , y ) בסביבה של ( x 0 , y 0 ) (x_{0},y_{0}) ( x 0 , y 0 ) .
הוכיחו כי ∇ ⃗ g ( x 0 , y 0 ) = ( 0 , 0 ) \vec{\nabla}g(x_{0},y_{0})=(0,0) ∇ g ( x 0 , y 0 ) = ( 0 , 0 ) .
הערה: זהו משפט פרמה עבור פונקציה בשני משתנים.
הוכחה:
נתון כי g ( x , y ) g(x,y) g ( x , y ) בעלת נ"ח רציפות ב-D D D ולכן, עבור ( x 0 , y 0 ) ∈ D (x_{0},y_{0})\in D ( x 0 , y 0 ) ∈ D מתקיים:
∂ g ∂ x ( x 0 , y 0 ) = lim h → 0 g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) h ∂ g ∂ y ( x 0 , y 0 ) = lim h → 0 g ( x 0 , y 0 + h ) − g ( x 0 , y 0 ) h \parder{g}{x}(x_{0},y_{0})=\lim_{ h \to 0 } \frac{g(x_{0}+h,y_{0})-g(x_{0},y_{0})}{h} \qquad \qquad \parder{g}{y}(x_{0},y_{0})=\lim_{ h \to 0 } \frac{g(x_{0},y_{0}+h)-g(x_{0},y_{0})}{h} ∂ x ∂ g ( x 0 , y 0 ) = h → 0 lim h g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) ∂ y ∂ g ( x 0 , y 0 ) = h → 0 lim h g ( x 0 , y 0 + h ) − g ( x 0 , y 0 )
נסתכל על ∂ g ∂ x \parder{g}{x} ∂ x ∂ g תחילה.
עבור h > 0 h>0 h > 0 מתקיים:
g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) h < ↓ g ( x 0 , y 0 ) ≥ g ( x 0 + h , y 0 ) 0 \frac{g(x_{0}+h,y_{0})-g(x_{0},y_{0})}{h} \underset{\begin{matrix}
\downarrow \\
g(x_{0},y_{0})\geq g(x_{0}+h,y_{0})
\end{matrix}}{<} 0 h g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) ↓ g ( x 0 , y 0 ) ≥ g ( x 0 + h , y 0 ) < 0
ולכן, לפי יחס סדר גבולות:
lim h → 0 + g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) h ≤ 0 \lim_{ h \to 0^+ } \frac{g(x_{0}+h,y_{0})-g(x_{0},y_{0})}{h}\leq 0 h → 0 + lim h g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) ≤ 0
עבור h < 0 h<0 h < 0 מתקיים:
g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) h > ↓ g ( x 0 , y 0 ) ≥ g ( x 0 + h , y 0 ) 0 \frac{g(x_{0}+h,y_{0})-g(x_{0},y_{0})}{h} \underset{\begin{matrix}
\downarrow \\
g(x_{0},y_{0})\geq g(x_{0}+h,y_{0})
\end{matrix}}{>} 0 h g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) ↓ g ( x 0 , y 0 ) ≥ g ( x 0 + h , y 0 ) > 0
ולכן, לפי יחס סדר גבולות:
lim h → 0 − g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) h ≥ 0 \lim_{ h \to 0^- } \frac{g(x_{0}+h,y_{0})-g(x_{0},y_{0})}{h}\geq 0 h → 0 − lim h g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) ≥ 0
לכן, לפי יחידות הגבול וקיומו, אנו יודעים כי:
∂ g ∂ x ( x 0 , y 0 ) = lim h → 0 + g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) h = lim h → 0 − g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) h = 0 \parder{g}{x}(x_{0},y_{0})=\lim_{ h \to 0^+ } \frac{g(x_{0}+h,y_{0})-g(x_{0},y_{0})}{h}=\lim_{ h \to 0^- } \frac{g(x_{0}+h,y_{0})-g(x_{0},y_{0})}{h}=0 ∂ x ∂ g ( x 0 , y 0 ) = h → 0 + lim h g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) = h → 0 − lim h g ( x 0 + h , y 0 ) − g ( x 0 , y 0 ) = 0
באותו האופן נקבל כי:
∂ g ∂ y ( x 0 , y 0 ) = 0 \parder{g}{y}(x_{0},y_{0})=0 ∂ y ∂ g ( x 0 , y 0 ) = 0
לפי הגדרת הגרדיאנט מתקיים:
∇ g ( x 0 , y 0 ) = ( ∂ g ∂ x ( x 0 , y 0 ) ∂ g ∂ y ( x 0 , y 0 ) ) = ( 0 , 0 ) \nabla g(x_{0},y_{0})=\begin{pmatrix}
\displaystyle\parder{g}{x}(x_{0},y_{0}) & \displaystyle\parder{g}{y}(x_{0},y_{0})
\end{pmatrix}=\begin{pmatrix}
0, 0
\end{pmatrix} ∇ g ( x 0 , y 0 ) = ( ∂ x ∂ g ( x 0 , y 0 ) ∂ y ∂ g ( x 0 , y 0 ) ) = ( 0 , 0 )
∎
תרגיל 2
עבור שתי נקודות p = ( x 1 , y 1 ) , q = ( x 2 , y 2 ) ∈ R 2 p=(x_{1},y_{1}),\;q=(x_{2},y_{2})\in \mathbb{R}^2 p = ( x 1 , y 1 ) , q = ( x 2 , y 2 ) ∈ R 2 נסמן [ p , q ] [p,q] [ p , q ] את הקטע הישר המחבר בינהן, כלומר, קבוצת הנקודות { ( 1 − t ) p + t q ∣ 0 ≤ t ≤ 1 } \left\{ (1-t)p+tq\;|\;0\leq t\leq 1 \right\} { ( 1 − t ) p + tq ∣ 0 ≤ t ≤ 1 } .
תהי f ( x , y ) f(x,y) f ( x , y ) פונקציה גזירה בתחום פתוח וקשיר D D D .
הוכיחו שלכל p , q ∈ D p,q \in D p , q ∈ D כך ש-[ p , q ] ⊂ D [p,q]\subset D [ p , q ] ⊂ D קיימת s ∈ [ p , q ] s \in[p,q] s ∈ [ p , q ] כך שמתקיים:
f ( q ) − f ( p ) = ∇ f ( s ) ⋅ ( q − p ) f(q)-f(p)=\nabla f(s)\cdot(q-p) f ( q ) − f ( p ) = ∇ f ( s ) ⋅ ( q − p )
הערה: זוהי הכללה של משפט לגרנז' עבור פונקציית שני משתנים.
הדרכה: הגדירו g ( t ) = ( 1 − t ) p + t q g(t)=(1-t)p+tq g ( t ) = ( 1 − t ) p + tq והסתכלו על φ ( t ) = f ( g ( t ) ) \varphi(t)=f(g(t)) φ ( t ) = f ( g ( t )) .
הוכחה:
נגדיר g ( t ) = ( 1 − t ) p + t q g(t)=(1-t)p+tq g ( t ) = ( 1 − t ) p + tq ונסמן φ ( t ) = f ( g ( t ) ) \varphi(t)=f(g(t)) φ ( t ) = f ( g ( t )) .
נשים לב כי g ( t ) = ( x ( t ) , y ( t ) ) g(t)=(x(t),y(t)) g ( t ) = ( x ( t ) , y ( t )) כאשר x ( t ) = ( 1 − t ) x 1 + t x 2 x(t)=(1-t)x_{1}+tx_{2} x ( t ) = ( 1 − t ) x 1 + t x 2 ו-y ( t ) = ( 1 − t ) y 1 + t y 2 y(t)=(1-t)y_{1}+ty_{2} y ( t ) = ( 1 − t ) y 1 + t y 2 ואז, לפי כלל השרשרת(כי f f f גזירה ב-D D D ו-x ( t ) , y ( t ) x(t),y(t) x ( t ) , y ( t ) גזירות ב-[ 0 , 1 ] [0,1] [ 0 , 1 ] כאשר לכל t ∈ [ 0 , 1 ] t \in[0,1] t ∈ [ 0 , 1 ] מתקיים x ( t ) , y ( t ) ∈ D x(t),y(t)\in D x ( t ) , y ( t ) ∈ D שכן [ p , q ] ⊂ D [p,q]\subset D [ p , q ] ⊂ D )
φ ( t ) = f ( g ( t ) ) ⟹ φ ′ ( t ) = ∂ f ∂ x ( x ( t ) , y ( t ) ) ⋅ d x d t + ∂ f ∂ y ( x ( t ) , y ( t ) ) ⋅ d y d t \varphi(t)=f(g(t))\implies \varphi'(t)=\parder{f}{x}(x(t),y(t))\cdot \frac{dx}{dt}+\parder{f}{y}(x(t),y(t))\cdot \frac{dy}{dt} φ ( t ) = f ( g ( t )) ⟹ φ ′ ( t ) = ∂ x ∂ f ( x ( t ) , y ( t )) ⋅ d t d x + ∂ y ∂ f ( x ( t ) , y ( t )) ⋅ d t d y
נשים לב כי x ′ ( t ) = − x 1 + x 2 = x 2 − x 1 x'(t)=-x_{1}+x_{2}=x_{2}-x_{1} x ′ ( t ) = − x 1 + x 2 = x 2 − x 1 וכי y ′ ( t ) = − y 1 + y 2 = y 2 − y 1 y'(t)=-y_{1}+y_{2}=y_{2}-y_{1} y ′ ( t ) = − y 1 + y 2 = y 2 − y 1 ולכן:
φ ′ ( t ) = ∂ f ∂ x ( x ( t ) , y ( t ) ) ⋅ ( x 2 − x 1 ) + ∂ f ∂ y ( x ( t ) , y ( t ) ) ⋅ ( y 2 − y 1 ) \varphi'(t)=\parder{f}{x}(x(t),y(t))\cdot (x_{2}-x_{1})+\parder{f}{y}(x(t),y(t))\cdot (y_{2}-y_{1}) φ ′ ( t ) = ∂ x ∂ f ( x ( t ) , y ( t )) ⋅ ( x 2 − x 1 ) + ∂ y ∂ f ( x ( t ) , y ( t )) ⋅ ( y 2 − y 1 )
הפונקציה φ ( t ) \varphi(t) φ ( t ) רציפה בקטע [ 0 , 1 ] [0,1] [ 0 , 1 ] מהרכבה של רציפות וגזירה ב-( 0 , 1 ) (0,1) ( 0 , 1 ) מהרכבה של גזירות ולכן, לפי לגרנז', קיים t 0 ∈ ( 0 , 1 ) t_{0} \in(0,1) t 0 ∈ ( 0 , 1 ) כך שעבורו:
φ ′ ( t 0 ) = φ ( 1 ) − φ ( 0 ) = f ( g ( 1 ) ) − f ( g ( 0 ) ) = f ( q ) − f ( p ) \varphi'(t_{0})=\varphi(1)-\varphi(0)=f(g(1))-f(g(0))=f(q)-f(p) φ ′ ( t 0 ) = φ ( 1 ) − φ ( 0 ) = f ( g ( 1 )) − f ( g ( 0 )) = f ( q ) − f ( p )
מכיוון שזוהי נגזרת נוכל להגיד גם כי:
φ ′ ( t 0 ) = ∂ f ∂ x ( g ( t 0 ) ) ( x 2 − x 1 ) + ∂ f ∂ y ( g ( t 0 ) ) ( y 2 − y 1 ) = ↓ s = g ( t 0 ) ∂ f ∂ x ( s ) ( x 2 − x 1 ) + ∂ f ∂ y ( s ) ( y 2 − y 1 ) \varphi'(t_{0})=\parder{f}{x}(g(t_{0}))(x_{2}-x_{1})+\parder{f}{y}(g(t_{0}))(y_{2}-y_{1}) \underset{\begin{matrix}
\downarrow \\
s=g(t_{0})
\end{matrix}}{=} \parder{f}{x}(s)(x_{2}-x_{1})+\parder{f}{y}(s)(y_{2}-y_{1}) φ ′ ( t 0 ) = ∂ x ∂ f ( g ( t 0 )) ( x 2 − x 1 ) + ∂ y ∂ f ( g ( t 0 )) ( y 2 − y 1 ) ↓ s = g ( t 0 ) = ∂ x ∂ f ( s ) ( x 2 − x 1 ) + ∂ y ∂ f ( s ) ( y 2 − y 1 )
נשים לב כי לפי הגדרת מכפלה סקלרית:
∇ f ( s ) ⋅ ( q − p ) = ( ∂ f ∂ x ( s ) ∂ f ∂ y ( s ) ) ⋅ ( x 2 − x 1 y 2 − y 1 ) = ∂ f ∂ x ( s ) ( x 2 − x 1 ) + ∂ f ∂ y ( s ) ( y 2 − y 1 ) \nabla f(s)\cdot (q-p)=\begin{pmatrix}
\parder{f}{x}(s) & \parder{f}{y}(s)
\end{pmatrix}\cdot \begin{pmatrix}
x_{2}-x_{1} & y_{2}-y_{1}
\end{pmatrix}=\parder{f}{x}(s)(x_{2}-x_{1})+\parder{f}{y}(s)(y_{2}-y_{1}) ∇ f ( s ) ⋅ ( q − p ) = ( ∂ x ∂ f ( s ) ∂ y ∂ f ( s ) ) ⋅ ( x 2 − x 1 y 2 − y 1 ) = ∂ x ∂ f ( s ) ( x 2 − x 1 ) + ∂ y ∂ f ( s ) ( y 2 − y 1 )
ולכן מתקיים:
f ( q ) − f ( p ) = ∂ f ∂ x ( s ) ( x 2 − x 1 ) + ∂ f ∂ y ( s ) ( y 2 − y 1 ) f(q)-f(p)=\parder{f}{x}(s)(x_{2}-x_{1})+\parder{f}{y}(s)(y_{2}-y_{1}) f ( q ) − f ( p ) = ∂ x ∂ f ( s ) ( x 2 − x 1 ) + ∂ y ∂ f ( s ) ( y 2 − y 1 )
∎
תרגיל 3
חשבו:
∫ 0 1 x n ln x d x \int_{0}^1 x^n\ln x\,dx 0 ∫ 1 x n ln x d x
לכל n ∈ N n \in \mathbb{N} n ∈ N באמצעות כלל לייבניץ.
הדרכה: לכל n ∈ N n \in \mathbb{N} n ∈ N התבוננו בפונקציה f ( x , t ) = x t f(x,t)=x^t f ( x , t ) = x t המוגדרת במלבן D = [ 0 , 1 ] × [ n 2 , 3 2 n ] D=[0,1]\times\left[ \frac{n}{2},\; \frac{3}{2}n \right] D = [ 0 , 1 ] × [ 2 n , 2 3 n ] .
נביט בפונקציה f ( x , t ) = x t f(x,t)=x^t f ( x , t ) = x t המוגדרת ב-D = [ 0 , 1 ] × [ n 2 , 3 n 2 ] D=[0,1]\times\left[ \frac{n}{2}, \frac{3n}{2} \right] D = [ 0 , 1 ] × [ 2 n , 2 3 n ] ונשים לב כי לפי כללי גזירה מתקיים:
∂ f ∂ t ( x , t ) = x t ⋅ ln x \parder{f}{t}(x,t)=x^t\cdot \ln x ∂ t ∂ f ( x , t ) = x t ⋅ ln x
לאור היותה של הנ"ח אלמנטרית, היא רציפה ב-D D D ולכן, ניתן להשתמש במשפט לייבניץ (כי האינטגרל מוגדר)
אזי, מתקיים:
F ( t ) = ∫ 0 1 x n ln x d x = ∫ 0 1 ∂ f ∂ t ( x , t ) d x F(t)=\int_{0}^1 x^n\ln x\;dx=\int_{0}^1 \parder{f}{t}(x,t)dx F ( t ) = 0 ∫ 1 x n ln x d x = 0 ∫ 1 ∂ t ∂ f ( x , t ) d x
לפי לייבניץ מתקיים:
F ( t ) = d d t ∫ 0 1 x t d x = ↓ N L d d t ( x t + 1 t + 1 ∣ 0 1 ) = d d t ( 1 t + 1 ) = − 1 ( t + 1 ) 2 F(t)=\frac{d}{dt}\int_{0}^1 x^tdx\underset{\begin{matrix}
\downarrow \\
NL
\end{matrix}}{=}\frac{d}{dt}\left(\left. \frac{x^{t+1}}{t+1} \right\vert_{0}^1\right)=\frac{d}{dt}\left( \frac{1}{t+1} \right)=\frac{-1}{(t+1)^2} F ( t ) = d t d 0 ∫ 1 x t d x ↓ N L = d t d ( t + 1 x t + 1 0 1 ) = d t d ( t + 1 1 ) = ( t + 1 ) 2 − 1
לכן, לכל n ∈ N n \in \mathbb{N} n ∈ N מתקיים:
∫ 0 1 x n ln x d x = − 1 ( n + 1 ) 2 \int_{0}^1 x^n\ln x\,dx=-\frac{1}{(n+1)^2} 0 ∫ 1 x n ln x d x = − ( n + 1 ) 2 1
לכל n , m ∈ N n,m \in \mathbb{N} n , m ∈ N חשבו באמצעות כלל לייבניץ את:
∫ 0 1 x n ln m x d x \int_{0}^1 x^n\ln^mx\,dx 0 ∫ 1 x n ln m x d x
רמז: ניתן לגזור m m m פעמים.
נסתכל על הפונקציה f ( x , t ) = x t f(x,t)=x^t f ( x , t ) = x t המוגדרת ב-D = [ 0 , 1 ] × [ n 2 , 3 n 2 ] D=[0,1]\times\left[ \frac{n}{2}, \frac{3n}{2} \right] D = [ 0 , 1 ] × [ 2 n , 2 3 n ] ונשים לב כי לפי כללי גזירה מתקיים:
∂ m f ∂ t m = x t ⋅ ln m x \nparder{m}{f}{t}=x^t\cdot \ln^mx ∂ t m ∂ m f = x t ⋅ ln m x
לאור היותה של הנ"ח אלמנטרית, היא רציפה ב-D D D ולכן, ניתן להשתמש במשפט לייבניץ (כי האינטגרל מוגדר)
אזי מתקיים:
F ( t ) = ∫ 0 1 x t ln m x d x = ∫ 0 1 ∂ m f ∂ t m ( x , t ) d x = d m d t m ∫ 0 1 x t d x = ↓ N L d m d t m ( x t + 1 t + 1 ∣ 0 1 ) = d m d t m 1 t + 1 F(t)=\int_{0}^1 x^t\ln^m x\,dx=\int_{0}^1 \nparder{m}{f}{t} (x,t)dx=\frac{d^m}{dt^m}\int_{0}^1 x^t dx \underset{\begin{matrix}
\downarrow \\
NL
\end{matrix}}{=}\frac{d^m} {dt^m}\left( \left. \frac{x^{t+1}}{t+1} \right\vert_{0}^1 \right)=\frac{d^m}{dt^m} \frac{1}{t+1} F ( t ) = 0 ∫ 1 x t ln m x d x = 0 ∫ 1 ∂ t m ∂ m f ( x , t ) d x = d t m d m 0 ∫ 1 x t d x ↓ N L = d t m d m ( t + 1 x t + 1 0 1 ) = d t m d m t + 1 1
כעת, לפי חוקי גזירה נקבל כי:
F ( t ) = ( − 1 ) m ( t + 1 ) m + 1 F(t)=\frac{(-1)^{m}}{(t+1)^{m+1}} F ( t ) = ( t + 1 ) m + 1 ( − 1 ) m
ולכן:
∫ 0 1 x n ln m x d x = ( − 1 ) m ( n + 1 ) m + 1 \int_{0}^1 x^n\ln^mx\,dx=\frac{(-1)^m}{(n+1)^{m+1}} 0 ∫ 1 x n ln m x d x = ( n + 1 ) m + 1 ( − 1 ) m
לכל n , m ∈ N n,m \in \mathbb{N} n , m ∈ N .
תרגיל 4
תהי f : [ a , ∞ ) → R f:[a,\infty)\to \mathbb{R} f : [ a , ∞ ) → R רציפה. לכל n ∈ N n \in \mathbb{N} n ∈ N נגדיר:
I n ( x ) = ∫ a x ( x − t ) n − 1 f ( t ) d t I_{n}(x)=\int_{a}^x (x-t)^{n-1}f(t)dt I n ( x ) = a ∫ x ( x − t ) n − 1 f ( t ) d t
הוכיחו כי לכל n ∈ N n \in \mathbb{N} n ∈ N מתקיים:
d n I n d x n = ( n − 1 ) ! f ( x ) \frac{d^nI_{n}}{dx^n}=(n-1)!f(x) d x n d n I n = ( n − 1 )! f ( x )
הוכחה:
נוכיח באינוקציה על n ∈ N n \in \mathbb{N} n ∈ N :
בסיס: עבור n = 1 n=1 n = 1 , נתון כי f f f רציפה ולכן ( x − t ) f ( t ) (x-t)f(t) ( x − t ) f ( t ) רציפה גם כן ולכן, לפי המשפט היסודי:
I 1 ( x ) = ∫ a x ( x − t ) 0 f ( t ) d t = ∫ a x f ( t ) d t ⟹ d I 1 d x = f ( x ) = 1 ⋅ f ( x ) = 0 ! ⋅ f ( x ) = ( 1 − 1 ) ! f ( x ) I_{1}(x)=\int_{a}^x (x-t)^{0}f(t)dt=\int_{a}^x f(t)dt\implies \frac{dI_{1}}{dx}= f(x)= 1\cdot f(x)=0!\cdot f(x)=(1-1)!f(x) I 1 ( x ) = a ∫ x ( x − t ) 0 f ( t ) d t = a ∫ x f ( t ) d t ⟹ d x d I 1 = f ( x ) = 1 ⋅ f ( x ) = 0 ! ⋅ f ( x ) = ( 1 − 1 )! f ( x )
צעד: נניח כי עבור n ∈ N n \in \mathbb{N} n ∈ N מתקיים d n I n d x n = ( n − 1 ) ! f ( x ) \frac{d^nI_{n}}{dx^n}=(n-1)!f(x) d x n d n I n = ( n − 1 )! f ( x ) ונוכיח כי d n + 1 I n + 1 d x n + 1 = n ! f ( x ) \frac{d^{n+1}I_{n+1}}{dx^{n+1}}=n!f(x) d x n + 1 d n + 1 I n + 1 = n ! f ( x ) .
נגדיר g ( x , t ) = ( x − t ) n − 1 f ( t ) d t g(x,t)=(x-t)^{n-1}f(t)dt g ( x , t ) = ( x − t ) n − 1 f ( t ) d t ונגדיר b ( x ) = x b(x)=x b ( x ) = x אשר הינה פונקציה גזירה בכל R \mathbb{R} R ואז:
I n = ∫ a b ( x ) g ( x , t ) d t I_{n}=\int_{a}^{b(x)} g(x,t)dt I n = a ∫ b ( x ) g ( x , t ) d t
נשים לב כי g g g גזירה ברציפות תחת x x x שכן:
∂ g ∂ x ( x , t ) = ( n − 1 ) ( x − t ) n − 2 ⋅ 1 ⋅ f ( t ) = ( n − 1 ) ( x − t ) n − 2 f ( t ) \parder{g}{x}(x,t)=(n-1)(x-t)^{n-2}\cdot 1\cdot f(t)=(n-1)(x-t)^{n-2}f(t) ∂ x ∂ g ( x , t ) = ( n − 1 ) ( x − t ) n − 2 ⋅ 1 ⋅ f ( t ) = ( n − 1 ) ( x − t ) n − 2 f ( t )
היא פונקציה רציפה מכפל של רציפות וקבועות.
לכן, לפי ההכלה למשפט לייבניץ' מתקיים:
d I n + 1 d x = ∫ a x ∂ g ∂ x ( x , t ) d t + g ( b ( x ) , x ) ⋅ b ′ ( x ) = ∫ a x n ( x − t ) n − 1 f ( t ) d t + ( x − x ) n f ( x ) = ↓ מונוטוניות n ∫ a x ( x − t ) n − 1 f ( t ) d t \begin{align*}
\frac{dI_{n+1}}{dx}&=\int_{a}^x \parder{g}{x}(x,t)dt+g(b(x),x)\cdot b'(x)=\int_{a}^x n(x-t)^{n-1}f(t)dt+(x-x)^nf(x) \\
&\!\!\!\,\!\underset{\begin{matrix}
\downarrow \\
\R{\text{מונוטוניות}}
\end{matrix}}{=}\!\!\!\!\! n\int_{a}^x (x-t)^{n-1}f(t)dt
\end{align*} d x d I n + 1 = a ∫ x ∂ x ∂ g ( x , t ) d t + g ( b ( x ) , x ) ⋅ b ′ ( x ) = a ∫ x n ( x − t ) n − 1 f ( t ) d t + ( x − x ) n f ( x ) ↓ מונוטוניות = n a ∫ x ( x − t ) n − 1 f ( t ) d t
d n + 1 I n + 1 d x n + 1 = d n d x n ( d I n + 1 d x ) = d n d x n n ∫ a x ( x − t ) n − 1 f ( t ) d t = ↓ הנחת האינדוקציה n ( n − 1 ) ! f ( t ) = n ! f ( t ) \frac{d^{n+1}I_{n+1}}{dx^{n+1}}=\frac{d^n}{dx^n}\left( \frac{dI_{n+1}}{dx} \right)=\frac{d^n}{dx^n}n\int_{a}^x (x-t)^{n-1}f(t)dt \underset{\begin{matrix}
\downarrow \\
\R{\text{הנחת האינדוקציה}}
\end{matrix}}{=}n(n-1)!f(t)=n!f(t) d x n + 1 d n + 1 I n + 1 = d x n d n ( d x d I n + 1 ) = d x n d n n a ∫ x ( x − t ) n − 1 f ( t ) d t ↓ הנחת האינדוקציה = n ( n − 1 )! f ( t ) = n ! f ( t )
לכן, לכל n ∈ N n \in \mathbb{N} n ∈ N מתקיים:
d n I n d x n = ( n − 1 ) ! f ( t ) \frac{d^nI_{n}}{dx^n}=(n-1)!f(t) d x n d n I n = ( n − 1 )! f ( t )
∎