\documentclass[a4paper,french]{article}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\usepackage{amsmath}
\usepackage{fancyhdr}
\usepackage[latin1]{inputenc}
\usepackage{theorem}
\usepackage{amssymb}
\usepackage{multicol}
\usepackage{hyperref}
\usepackage[a4paper]{geometry}
\setcounter{MaxMatrixCols}{10}
%TCIDATA{OutputFilter=LATEX.DLL}
%TCIDATA{Version=5.00.0.2552}
%TCIDATA{}
%TCIDATA{Created=Monday, June 19, 2006 20:45:35}
%TCIDATA{LastRevised=Friday, June 23, 2006 18:39:57}
%TCIDATA{}
%TCIDATA{}
%TCIDATA{CSTFile=40 LaTeX article.cst}
\geometry{margin={1cm,2cm}}
\pagestyle{fancy}
\lhead{PHEC1}
\chead{Correction ds n°7}
\rhead{2005-2006}
\cfoot{\thepage/\pageref{fin}}
\lfoot{\href{http://abdellah.bechata.free.fr}{www.mathematiques.fr.st}}
\rfoot{abdellah bechata}
\theoremstyle{break}
\newtheorem{theorem}{Théorème}
\newtheorem{proposition}[theorem]{Proposition}
\newtheorem{lemma}[theorem]{Lemme}
\newtheorem{corollary}[theorem]{Corollaire}
\theorembodyfont{\rmfamily}
\newtheorem{correction}[theorem]{correction de l'exercice}
\newtheorem{acknowledgement}[theorem]{Prérequis}
\newtheorem{algorithm}[theorem]{Algorithme}
\newtheorem{axiom}[theorem]{Axiome}
\newtheorem{case}[theorem]{Cas}
\newtheorem{claim}[theorem]{Affirmation}
\newtheorem{conclusion}[theorem]{Conclusion}
\newtheorem{condition}[theorem]{Condition}
\newtheorem{conjecture}[theorem]{Conjecture}
\newtheorem{criterion}[theorem]{Critèrion}
\newtheorem{definition}[theorem]{Définition}
\newtheorem{example}[theorem]{Exemple}
\newtheorem{exercise}[theorem]{Exercice}
\newtheorem{notation}[theorem]{Notation}
\newtheorem{problem}[theorem]{Problème}
\newtheorem{remark}[theorem]{Remarque}
\newtheorem{solution}[theorem]{Solution}
\newtheorem{summary}[theorem]{Sommaire}
\newenvironment{proof}[1][Preuve]{\noindent\textbf{#1.} }{\ \rule{0.5em}{0.5em}}
\setlength{\columnseprule}{0.5pt}
\input{tcilatex}
\begin{document}
\section*{CORRECTION PROBLEME}
\subsection*{I. Structure de $E$.}
\begin{enumerate}
\item Etant donné que l'on demande également une base, on va directement
montrer que $E$ est engendré linéairement par certains vecteurs.\newline
$X\in E$ si et seulement il existe deux réels $a$ et $b$ tels que
\begin{eqnarray*}
X &\in &E\Leftrightarrow \text{ il existe deux réels }a,b\text{ tels que }%
X=M(a,b)=%
\begin{pmatrix}
b & a & b \\
a & b & b \\
b & b & a%
\end{pmatrix}%
=a%
\begin{pmatrix}
0 & 1 & 0 \\
1 & 0 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
+b%
\begin{pmatrix}
1 & 0 & 0 \\
0 & 1 & 1 \\
1 & 1 & 0%
\end{pmatrix}
\\
&\Leftrightarrow &X\in \func{Vect}\left\{
\begin{pmatrix}
0 & 1 & 0 \\
1 & 0 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
,%
\begin{pmatrix}
1 & 0 & 0 \\
0 & 1 & 1 \\
1 & 1 & 0%
\end{pmatrix}%
\right\} \Rightarrow E=\func{Vect}\left\{
\begin{pmatrix}
0 & 1 & 0 \\
1 & 0 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
,%
\begin{pmatrix}
1 & 0 & 0 \\
0 & 1 & 1 \\
1 & 1 & 0%
\end{pmatrix}%
\right\}
\end{eqnarray*}%
Par conséquent, $E$ est un espace vectoriel
\item La famille $\left\{
\begin{pmatrix}
0 & 1 & 0 \\
1 & 0 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
,%
\begin{pmatrix}
1 & 0 & 0 \\
0 & 1 & 1 \\
1 & 1 & 0%
\end{pmatrix}%
\right\} $ est génératrice de $E.$ Montrons qu'il s'agit d'une famille libre
de $E.$ Soient $a$ et $b$ deux réels tels que
\begin{equation*}
a%
\begin{pmatrix}
0 & 1 & 0 \\
1 & 0 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
+b%
\begin{pmatrix}
1 & 0 & 0 \\
0 & 1 & 1 \\
1 & 1 & 0%
\end{pmatrix}%
=0_{\mathfrak{M}_{3}(\mathbb{R})}=%
\begin{pmatrix}
0 & 0 & 0 \\
0 & 0 & 0 \\
0 & 0 & 0%
\end{pmatrix}%
\Leftrightarrow
\begin{pmatrix}
b & a & b \\
a & b & b \\
b & b & a%
\end{pmatrix}%
=%
\begin{pmatrix}
0 & 0 & 0 \\
0 & 0 & 0 \\
0 & 0 & 0%
\end{pmatrix}%
\Leftrightarrow a=b=0
\end{equation*}%
Par conséquent, la famille $\left\{
\begin{pmatrix}
0 & 1 & 0 \\
1 & 0 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
,%
\begin{pmatrix}
1 & 0 & 0 \\
0 & 1 & 1 \\
1 & 1 & 0%
\end{pmatrix}%
\right\} $ est bien une base de $E.$
\end{enumerate}
\subsection*{II. Etude d'une certaine application linéaire.}
\begin{enumerate}
\item Soient $\alpha ,\beta $ deux réels et $X=(x,y,z)$ et $Y=(a,b,c)$ deux
vecteurs de $\mathfrak{M}_{1,3}(\mathbb{R})$%
\begin{eqnarray*}
f(\alpha X+\beta Y) &=&f(\alpha (x,y,z)+\beta (a,b,c))=f(\alpha x+\beta
a,\alpha y+\beta b,\alpha z+\beta c) \\
&=&(\alpha y+\beta b,\alpha x+\beta a,\alpha z+\beta c)\quad ("x=\alpha
x+\beta a,\quad y=\alpha y+\beta b,\quad z=\alpha z+\beta c") \\
&=&\alpha (y,x,z)+\beta (b,a,c)=\alpha f(X)+\beta f(Y)
\end{eqnarray*}
\item Il s'agit d'exprimer les images de chaque vecteur de la base de
l'espace de départ $(f(\varepsilon _{1}),f(\varepsilon _{2}),f(\varepsilon
_{3})$ ici) comme combinaison linéaire des vecteurs de la base de l'espace
d'arrivée ($\varepsilon _{1},\varepsilon _{2},\varepsilon _{3}$ ici) donc il
s'agit d'exprimer $f(\varepsilon _{1})$ (resp. $f(\varepsilon
_{2}),f(\varepsilon _{2}))$ comme combinaison linéaire de $\varepsilon
_{1},\varepsilon _{2},\varepsilon _{3}.$ Un calcul direct nous donne
\begin{eqnarray*}
f(\varepsilon _{1}) &=&f(1,0,0)=(0,1,0)=\varepsilon _{2}=0\times \varepsilon
_{1}+1\times \varepsilon _{2}+0\times \varepsilon _{3} \\
f(\varepsilon _{2}) &=&f(0,1,0)=(1,0,0)=\varepsilon _{1}=1\times \varepsilon
_{1}+0\times \varepsilon _{2}+0\times \varepsilon _{3} \\
f(\varepsilon _{3}) &=&f(0,0,1)=(0,0,1)=\varepsilon _{3}=0\times \varepsilon
_{1}+0\times \varepsilon _{2}+1\times \varepsilon _{3}
\end{eqnarray*}%
ce qui entraine%
\begin{equation*}
A=\overset{%
\begin{array}{ccc}
f(\varepsilon _{1}) & f(\varepsilon _{2}) & f(\varepsilon _{3})%
\end{array}%
}{%
\begin{pmatrix}
& 0 & & 1 & & 0 & \\
& 1 & & 0 & & 0 & \\
& 0 & & 0 & & 1 &
\end{pmatrix}%
}%
\begin{array}{c}
\varepsilon _{1} \\
\varepsilon _{2} \\
\varepsilon _{3}%
\end{array}%
\end{equation*}
\item Il est immédiat que $A=M(1,0)$.
\item Etant donné que l'on demande également une base, on va directement
montrer que $E$ est engendré linéairement par certains vecteurs.%
\begin{eqnarray*}
X &\in &E_{-1}(f)\Leftrightarrow X=(a,b,c)\text{ et }f(X)=-X\Leftrightarrow
(b,a,c)=-(a,b,c)\Leftrightarrow \left\{
\begin{array}{c}
b=-a \\
a=-b \\
c=-c%
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
a+b=0 \\
a+b=0 \\
2c=0%
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
a=-b \\
c=0%
\end{array}%
\right. \\
&\Leftrightarrow &X=(-b,b,0)=b(-1,1,0)\Leftrightarrow E_{-1}(f)=\func{Vect}%
((-1,1,0))
\end{eqnarray*}%
Par conséquent, $E_{-1}(f)$ est un espace vectoriel (comme espace engendré
par certains vecteurs) et $e_{1}=(-1,1,0)$ en est une base (unique vecteur gé%
nérateur et il est non nul)
\item On procède de la même façon%
\begin{eqnarray*}
X &\in &E_{-1}(f)\Leftrightarrow X=(a,b,c)\text{ et }f(X)=X\Leftrightarrow
(b,a,c)=(a,b,c)\Leftrightarrow \left\{
\begin{array}{c}
b=a \\
a=b \\
c=c%
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
-a+b=0 \\
a-b=0 \\
0=0%
\end{array}%
\right. \\
&\Leftrightarrow &\left\{
\begin{array}{c}
-a+b=0 \\
0=0 \\
0=0%
\end{array}%
\right. \left\vert
\begin{array}{c}
\text{«\emph{Pivot pour }}a\text{»} \\
L_{2}\leftarrow L_{2}+L_{1} \\
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
a=b%
\end{array}%
\right. \Leftrightarrow X=(b,b,c)=b(1,1,0)+c(0,0,1) \\
&\Leftrightarrow &E_{1}(f)=\func{Vect}((1,1,0),(0,0,1))
\end{eqnarray*}%
Par conséquent, $E_{1}(f)$ est un espace vectoriel (comme espace engendré
par certains vecteurs) et $\left\{ e_{2}=(1,1,0),e_{3}=(0,0,1)\right\} $ en
est une famille génératrice. Il reste à montrer qu'il s'agit d'une famille
libre. Soient $a$ et $b$ deux réels tels que
\begin{equation*}
ae_{2}+be_{3}=0_{\mathfrak{M}_{1,3}(\mathbb{R})}\Leftrightarrow
a(1,1,0)+b(0,0,1)=(0,0,0)\Leftrightarrow (a,a,b)=(0,0,0)\Leftrightarrow a=b=0
\end{equation*}%
On en déduit que la famille $(e_{2},e_{3})$ est une base de $E_{1}(f).$
\item Commençons par montrer qu'il s'agit d'une famille libre. Soient $a,b,c$
trois réels tels que
\begin{gather*}
ae_{1}+be_{2}+ce_{3}=0_{\mathfrak{M}_{1,3}(\mathbb{R})}\Leftrightarrow
a(-1,1,0)+b(1,1,0)+c(0,0,1)=(0,0,0) \\
\Leftrightarrow (-a+b,a+b,c)=(0,0,0)\Leftrightarrow \left\{
\begin{array}{c}
-a+b=0 \\
a+b=0 \\
c=0%
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
-a+b=0 \\
2b=0 \\
c=0%
\end{array}%
\right. \left\vert
\begin{array}{c}
\text{«\emph{Pivot pour }}a\text{»} \\
L_{2}\leftarrow L_{2}+L_{1} \\
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
c=0 \\
b=0 \\
a=0%
\end{array}%
\right.
\end{gather*}%
donc la famille $(e_{1},e_{2},e_{3})$ est bien libre.\newline
Montrons qu'il s'agit d'une famille génératrice. Soit $X=(x,y,z)\in
\mathfrak{M}_{1,3}(\mathbb{R})$, existe-t-il trois réels $a,b,c$ tels que
\begin{equation*}
X=ae_{1}+be_{2}+ce_{3}\Leftrightarrow
(x,y,z)=a(-1,1,0)+b(1,1,0)+c(0,0,1)\Leftrightarrow \left\{
\begin{array}{c}
-a+b=x \\
a+b=y \\
c=z%
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
-a+b=x \\
2b=x+y \\
c=z%
\end{array}%
\right. \left\vert
\begin{array}{c}
\text{«\emph{Pivot pour }}a\text{»} \\
L_{2}\leftarrow L_{2}+L_{1} \\
\end{array}%
\right.
\end{equation*}%
Par conséquent, on en déduit l'existence de $c$ puis de $b$ et enfin celui
de $a,$ ce qui entraine que les réels $a,b,c$ existent bien. Ainsi, la
famille $(e_{1},e_{2},e_{3})$ est génératrice et libre donc c'est une base
de $\mathfrak{M}_{1,3}(\mathbb{R})$
\item Il s'agit d'exprimer les images de chaque vecteur de la base de
l'espace de départ $(f(e_{1}),f(e_{2}),f(e_{3})$ ici) comme combinaison liné%
aire des vecteurs de la base de l'espace d'arrivée ($e_{1},e_{2},e_{3}$ ici)
donc il s'agit d'exprimer $f(e_{1})$ (resp. $f(e_{2}),f(e_{2}))$ comme
combinaison linéaire de $e_{1},e_{2},e_{3}.$ Puisque $e_{1}$ appartient à $%
E_{-1}(f),$ on a $f(e_{1})=-e_{1}$ (il est solution de $f(X)=-X)$ et comme $%
e_{2},e_{3}$ appartiennent à $E_{1}(f),$ on en déduit que $f(e_{2})=e_{2}$
et $f(e_{3})=e_{3},$ ce qui entraine%
\begin{equation*}
B=\overset{%
\begin{array}{ccc}
f(e_{1}) & f(e_{2}) & f(e_{3})%
\end{array}%
}{%
\begin{pmatrix}
& -1 & & 0 & & 0 & \\
& 0 & & 1 & & 0 & \\
& 0 & & 0 & & 1 &
\end{pmatrix}%
}%
\begin{array}{c}
e_{1} \\
e_{2} \\
e_{3}%
\end{array}%
\end{equation*}
\item On procède par les opérations élémentaires sur les matrices%
\begin{equation*}
\begin{pmatrix}
-1 & 1 & 0 \\
1 & 1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\quad \vdots \quad \left(
\begin{array}{ccc}
1 & 0 & 0 \\
0 & 1 & 0 \\
0 & 0 & 1%
\end{array}%
\right) \Leftrightarrow
\begin{pmatrix}
-1 & 1 & 0 \\
0 & 2 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\quad \vdots \quad \left(
\begin{array}{ccc}
1 & 0 & 0 \\
1 & 1 & 0 \\
0 & 0 & 1%
\end{array}%
\right) \left\vert
\begin{array}{c}
\text{«\emph{Pivot}» } \\
L_{2}\leftarrow L_{2}+L_{1} \\
\end{array}%
\right.
\end{equation*}%
La nouvelle matrice est triangulaire supérieure et tous ses coefficients
diagonaux sont non nuls donc elle est inversible, ce qui implique
l'inversibilité de $P.$ On poursuit en effectuant le pivot inversé.%
\begin{equation*}
\begin{pmatrix}
-2 & 0 & 0 \\
0 & 2 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\quad \vdots \quad \left(
\begin{array}{ccc}
1 & -1 & 0 \\
1 & 1 & 0 \\
0 & 0 & 1%
\end{array}%
\right) \left\vert
\begin{array}{c}
L_{1}\leftarrow 2L_{1}-L_{2} \\
\\
\end{array}%
\right. \Leftrightarrow
\begin{pmatrix}
-2 & 0 & 0 \\
0 & 2 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\quad \vdots \quad \left(
\begin{array}{ccc}
-\dfrac{1}{2} & \dfrac{1}{2} & 0\smallskip \\
\dfrac{1}{2} & \dfrac{1}{2} & 0\smallskip \\
0 & 0 & 1%
\end{array}%
\right) \left\vert
\begin{array}{c}
L_{1}\leftarrow -\dfrac{1}{2}L_{1}\smallskip \\
L_{2}\leftarrow -\dfrac{1}{2}L_{2}\smallskip \\
\end{array}%
\right.
\end{equation*}%
Par conséquent, on a $P^{-1}=\left(
\begin{array}{ccc}
-\dfrac{1}{2} & \dfrac{1}{2} & 0\smallskip \\
\dfrac{1}{2} & \dfrac{1}{2} & 0\smallskip \\
0 & 0 & 1%
\end{array}%
\right) .$ \emph{Vérification} : $%
\begin{pmatrix}
-1 & 1 & 0 \\
1 & 1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\left(
\begin{array}{ccc}
-\dfrac{1}{2} & \dfrac{1}{2} & 0\smallskip \\
\dfrac{1}{2} & \dfrac{1}{2} & 0\smallskip \\
0 & 0 & 1%
\end{array}%
\right) =\left(
\begin{array}{ccc}
1 & 0 & 0 \\
0 & 1 & 0 \\
0 & 0 & 1%
\end{array}%
\right) =I_{3}$\newline
Un calcul direct nous donne $PB=%
\begin{pmatrix}
1 & 1 & 0 \\
-1 & 1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\Rightarrow PBP^{-1}=\left(
\begin{array}{ccc}
0 & 1 & 0 \\
1 & 0 & 0 \\
0 & 0 & 1%
\end{array}%
\right) =A.$
\end{enumerate}
\subsection*{III. Calcul de $[M(a,b)]^{n}$}
\begin{enumerate}
\item Un calcul direct donne $QR=6I_{3}$ donc $Q\left( \dfrac{1}{6}R\right)
=I_{3},$ ce qui entraine que $Q$ est inversible et $Q^{-1}=\dfrac{1}{6}R$.
\item Un calcul direct donne
\begin{eqnarray*}
Q^{-1}M(a,b) &=&\dfrac{1}{6}RM(a,b)=\dfrac{1}{6}%
\begin{pmatrix}
2 & 2 & 2 \\
3 & -3 & 0 \\
1 & 1 & -2%
\end{pmatrix}%
\begin{pmatrix}
b & a & b \\
a & b & b \\
b & b & a%
\end{pmatrix}%
=\dfrac{1}{6}%
\begin{pmatrix}
2a+4b & 2a+4b & 2a+4b \\
-3a+3b & 3a-3b & 0 \\
a-b & a-b & -2a+2b%
\end{pmatrix}
\\
Q^{-1}M(a,b)Q &=&\dfrac{1}{6}%
\begin{pmatrix}
6a+12b & 0 & 0 \\
0 & -6a+6b & 0 \\
0 & 0 & 6a-6b%
\end{pmatrix}%
=%
\begin{pmatrix}
a+2b & 0 & 0 \\
0 & -a+b & 0 \\
0 & 0 & a-b%
\end{pmatrix}%
\end{eqnarray*}
\item Une récurrence archi-classique, que le lecteur fera, montre la formule
demandée.
\end{enumerate}
\subsection*{IV. Une application probabiliste.}
\begin{enumerate}
\item L'achat une année donnée dépendant de l'achat effectué l'année précé%
dente, nous incite à introduire les évènements $T_{n}$ : «acheter un jouet
traditionnel l'année $n$», $M_{n}$ : «acheter un jouet à la mode l'année $n$»%
, $S_{n}$ : «acheter un jouet scientifique l'année $n$». On applique la
formule des probabilités totales au système complet d'évènements $%
T_{n},M_{n},S_{n},$ ce qui nous donne
\begin{eqnarray*}
p_{n+1} &=&P(T_{n+1})=P(T_{n}\cap T_{n+1})+P(M_{n}\cap T_{n+1})+P(S_{n}\cap
T_{n+1}) \\
&=&P(T_{n})P_{(T_{n})}(T_{n+1})+P(M_{n})P_{(M_{n})}(T_{n+1})+P(S_{n})P_{(S_{n})}(T_{n+1})
\\
&=&\dfrac{1}{4}P(T_{n})+\dfrac{1}{2}P(M_{n})+\dfrac{1}{4}P(S_{n})=\dfrac{1}{4%
}p_{n}+\dfrac{1}{2}q_{n}+\dfrac{1}{4}r_{n} \\
q_{n+1} &=&P(M_{n+1})=P(T_{n}\cap M_{n+1})+P(M_{n}\cap M_{n+1})+P(S_{n}\cap
M_{n+1}) \\
&=&P(T_{n})P_{(T_{n})}(M_{n+1})+P(M_{n})P_{(M_{n})}(M_{n+1})+P(S_{n})P_{(S_{n})}(M_{n+1})
\\
&=&\dfrac{1}{2}P(T_{n})+\dfrac{1}{4}P(M_{n})+\dfrac{1}{4}P(S_{n})=\dfrac{1}{2%
}p_{n}+\dfrac{1}{4}q_{n}+\dfrac{1}{4}r_{n} \\
r_{n+1} &=&P(S_{n+1})=P(T_{n}\cap S_{n+1})+P(M_{n}\cap S_{n+1})+P(S_{n}\cap
S_{n+1}) \\
&=&P(T_{n})P_{(T_{n})}(S_{n+1})+P(M_{n})P_{(M_{n})}(S_{n+1})+P(S_{n})P_{(S_{n})}(S_{n+1})
\\
&=&\dfrac{1}{4}P(T_{n})+\dfrac{1}{4}P(M_{n})+\dfrac{1}{2}P(S_{n})=\dfrac{1}{4%
}p_{n}+\dfrac{1}{4}q_{n}+\dfrac{1}{2}r_{n}
\end{eqnarray*}
\item Il est immédiat que $%
\begin{pmatrix}
p_{n+1} \\
q_{n+1} \\
r_{n+1}%
\end{pmatrix}%
=\left(
\begin{array}{ccc}
\dfrac{1}{4} & \dfrac{1}{2} & \dfrac{1}{4}\smallskip \\
\dfrac{1}{2} & \dfrac{1}{4} & \dfrac{1}{4}\smallskip \\
\dfrac{1}{4} & \dfrac{1}{4} & \dfrac{1}{2}\smallskip%
\end{array}%
\right)
\begin{pmatrix}
p_{n} \\
q_{n} \\
r_{n}%
\end{pmatrix}%
=M\left( \dfrac{1}{2},\dfrac{1}{4}\right)
\begin{pmatrix}
p_{n} \\
q_{n} \\
r_{n}%
\end{pmatrix}%
$.
\item Une récurrence archi-classique, que le lecteur fera, montre la formule
demandée.
\item La question précédente combinée aux questions III.3 et III.\ 1, nous
donne%
\begin{eqnarray*}
\begin{pmatrix}
p_{n} \\
q_{n} \\
r_{n}%
\end{pmatrix}
&=&\dfrac{1}{6}\underbrace{%
\begin{pmatrix}
1 & 1 & 1 \\
1 & -1 & 1 \\
1 & 0 & -2%
\end{pmatrix}%
\begin{pmatrix}
1 & 0 & 0\smallskip \\
0 & \left( -\dfrac{1}{4}\right) ^{n} & 0\smallskip \\
0 & 0 & \left( \dfrac{1}{4}\right) ^{n}\smallskip%
\end{pmatrix}%
}\underbrace{%
\begin{pmatrix}
2 & 2 & 2 \\
3 & -3 & 0 \\
1 & 1 & -2%
\end{pmatrix}%
\begin{pmatrix}
1 \\
0 \\
0%
\end{pmatrix}%
}=\dfrac{1}{6}%
\begin{pmatrix}
1 & \left( -\dfrac{1}{4}\right) ^{n} & \left( \dfrac{1}{4}\right)
^{n}\smallskip \\
1 & -\left( -\dfrac{1}{4}\right) ^{n} & \left( \dfrac{1}{4}\right)
^{n}\smallskip \\
1 & 0 & -2\left( \dfrac{1}{4}\right) ^{n}\smallskip%
\end{pmatrix}%
\begin{pmatrix}
2 \\
3 \\
1%
\end{pmatrix}
\\
\begin{pmatrix}
p_{n} \\
q_{n} \\
r_{n}%
\end{pmatrix}
&=&\dfrac{1}{6}%
\begin{pmatrix}
2+3\left( -\dfrac{1}{4}\right) ^{n}+\left( \dfrac{1}{4}\right) ^{n}\smallskip
\\
2-3\left( -\dfrac{1}{4}\right) ^{n}+\left( \dfrac{1}{4}\right) ^{n}\smallskip
\\
-2\left( \dfrac{1}{4}\right) ^{n}+2\smallskip%
\end{pmatrix}%
=%
\begin{pmatrix}
\dfrac{1}{3}+\dfrac{1}{2}\left( -\dfrac{1}{4}\right) ^{n}+\dfrac{1}{6}\left(
\dfrac{1}{4}\right) ^{n} \\
\dfrac{1}{3}-\dfrac{1}{2}\left( -\dfrac{1}{4}\right) ^{n}+\dfrac{1}{6}\left(
\dfrac{1}{4}\right) ^{n} \\
\dfrac{1}{3}-\dfrac{1}{3}\left( \dfrac{1}{4}\right) ^{n}%
\end{pmatrix}%
\Rightarrow \left\{
\begin{array}{l}
p_{n}=\dfrac{1}{3}+\dfrac{1}{2}\left( -\dfrac{1}{4}\right) ^{n}+\dfrac{1}{6}%
\left( \dfrac{1}{4}\right) ^{n} \\
q_{n}=\dfrac{1}{3}-\dfrac{1}{2}\left( -\dfrac{1}{4}\right) ^{n}+\dfrac{1}{6}%
\left( \dfrac{1}{4}\right) ^{n} \\
r_{n}=\dfrac{1}{3}-\dfrac{1}{3}\left( \dfrac{1}{4}\right) ^{n}%
\end{array}%
\right.
\end{eqnarray*}
\item \thinspace
\begin{enumerate}
\item Par construction, on a $X_{n}(\Omega )=\{0,1\}$ et
\begin{eqnarray*}
P(X_{n} &=&1)=P(M_{n})=\dfrac{1}{3}-\dfrac{1}{2}\left( -\dfrac{1}{4}\right)
^{n}+\dfrac{1}{6}\left( \dfrac{1}{4}\right) ^{n}\quad
P(X_{n}=0)=1-P(X_{n}=1)=\dfrac{2}{3}+\dfrac{1}{2}\left( -\dfrac{1}{4}\right)
^{n}-\dfrac{1}{6}\left( \dfrac{1}{4}\right) ^{n} \\
E(X_{n}) &=&0\times P(X_{n}=0)+1\times P(X_{n}=1)=P(X_{n}=1)=\dfrac{2}{3}+%
\dfrac{1}{2}\left( -\dfrac{1}{4}\right) ^{n}-\dfrac{1}{6}\left( \dfrac{1}{4}%
\right) ^{n}
\end{eqnarray*}
\item On utilise la linéarité de l'espérance, ainsi que la formule $%
\dsum\limits_{k=1}^{n}q^{k}\underset{j=k-1}{=}\left(
\dsum\limits_{j=0}^{n-1}q^{j+1}\right)
=q\dsum\limits_{j=0}^{n-1}q^{j}=q\times \dfrac{1-q^{n}}{1-q}$%
\begin{eqnarray*}
E(Z_{n}) &=&E(X_{1}+X_{2}+\cdots +X_{n})=E(X_{1})+E(X_{2})+\cdots
+E(X_{n})=\dsum\limits_{k=1}^{n}\left[ \dfrac{1}{3}-\dfrac{1}{2}\left( -%
\dfrac{1}{4}\right) ^{k}+\dfrac{1}{6}\left( \dfrac{1}{4}\right) ^{k}\right]
\\
&=&\dsum\limits_{k=1}^{n}\dfrac{1}{3}-\dfrac{1}{2}\dsum\limits_{k=1}^{n}%
\left( -\dfrac{1}{4}\right) ^{k}+\dfrac{1}{6}\dsum\limits_{k=1}^{n}\left(
\dfrac{1}{4}\right) ^{k}=\dfrac{1}{3}n-\dfrac{1}{2}\left( -\dfrac{1}{4}%
\right) \dfrac{1-\left( -\dfrac{1}{4}\right) ^{n}}{1-\left( -\dfrac{1}{4}%
\right) }+\dfrac{1}{6}\left( \dfrac{1}{4}\right) \dfrac{1-\left( \dfrac{1}{4}%
\right) }{1-\dfrac{1}{4}} \\
&=&\dfrac{1}{3}n-\dfrac{1}{2}\left( -\dfrac{1}{4}\right) \left( \dfrac{4}{5}%
\right) \left[ 1-\left( -\dfrac{1}{4}\right) ^{n}\right] +\dfrac{1}{6}\left(
\dfrac{1}{4}\right) \left( \dfrac{4}{3}\right) \left[ 1-\left( \dfrac{1}{4}%
\right) ^{n}\right] \\
&=&\dfrac{1}{3}n+\dfrac{1}{10}\left[ 1-\left( -\dfrac{1}{4}\right) ^{n}%
\right] +\dfrac{1}{18}\left[ 1-\left( \dfrac{1}{4}\right) ^{n}\right] \\
\dfrac{E(Z_{n})}{n} &=&\dfrac{1}{3}+\dfrac{1}{10n}\left[ 1-\left( -\dfrac{1}{%
4}\right) ^{n}\right] +\dfrac{1}{18n}\left[ 1-\left( \dfrac{1}{4}\right) ^{n}%
\right] \underset{n\rightarrow +\infty }{\rightarrow }\dfrac{1}{3}\text{
(par les croissances comparées)}
\end{eqnarray*}
\item La variable $Z_{n}$ représente le nombre de jouets à la mode acheté
par un client durant des $n$ premières années\newline
$E(Z_{n})$ représente le nombre moyen de jouets à la mode acheté par un
client durant des $n$ premières années\newline
$\dfrac{E(Z_{n})}{n}$ représente le nombre moyen de jouets à la mode acheté
par année par un client durant les $n$ premières années.\newline
La limite obtenue à la question précédente montre que chaque client achète,
en moyenne, $\dfrac{1}{3}$ de jouets à la mode chaque année.
\end{enumerate}
\end{enumerate}
\subsection*{V. Etude des matrices commutant avec $M(2,1).$}
\begin{enumerate}
\item On procède par la caractérisation des sous-espaces vectoriels.\newline
$\mathcal{H}_{D}\subset \mathfrak{M}_{3}(\mathbb{R})$ qui est un espace
vectoriel et $\mathcal{H}_{D}\neq \emptyset $ puisqu'il contient $0_{3}$ (%
\emph{l'élément nul} de $\mathfrak{M}_{3}(\mathbb{R}))$ car $0_{3}\times
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
=0_{3}$%
\begin{equation*}
\left[
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
0_{3}=0_{3}\quad \text{et}\quad 0_{3}%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
=0_{3}\right] \Rightarrow
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
0_{3}=0_{3}%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\end{equation*}%
\textbf{Stabilité par combinaison linéaire} : Soient $X$ et $X^{\prime }$
deux éléments de $\mathcal{H}_{D}$ ainsi que deux réels $\lambda $ et $\mu .$
Il faut montrer que $\lambda X+\mu X^{\prime }\in \mathcal{H}_{D},$ c'est-à%
-dire que soient $X\in \mathfrak{M}_{3}(\mathbb{R})$ avec $%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
X=X%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
$ et $X^{\prime }\in \mathfrak{M}_{3}(\mathbb{R})$ avec $%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
X^{\prime }=X^{\prime }%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
,$ il faut montrer que $\lambda X+\mu X^{\prime }\in \mathfrak{M}_{3}(%
\mathbb{R})$ avec $%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
(\lambda X+\mu X^{\prime })=(\lambda X+\mu X^{\prime })%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
.$\newline
Pour cela, nous allons développer puis de regrouper les termes en $\lambda $
et les termes en $\mu $
\begin{eqnarray*}
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
(\lambda X+\mu X^{\prime }) &=&\lambda
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
X+\mu
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
X^{\prime } \\
&=&\lambda X%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
+\mu X^{\prime }%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
=(\lambda X+\mu X^{\prime })%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\end{eqnarray*}%
ce qui montre que $\mathcal{H}_{D}$ est stable par combinaison linéaire et $%
\mathcal{H}_{D}$ est bien un espace vectoriel.\newline
\textbf{Remarque} : \emph{nous n'utilisons pas les coordonnées car
l'ensemble }$\mathcal{H}_{D}$\emph{\ est défini par des relations sur l'élé%
ment }$X$\emph{\ directement.}
\item Soit $X=%
\begin{pmatrix}
a & b & c \\
d & e & f \\
g & h & i%
\end{pmatrix}%
\in \mathcal{H}_{D}$%
\begin{eqnarray*}
DX &=&XD\Leftrightarrow
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\begin{pmatrix}
a & b & c \\
d & e & f \\
g & h & i%
\end{pmatrix}%
=%
\begin{pmatrix}
a & b & c \\
d & e & f \\
g & h & i%
\end{pmatrix}%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\Leftrightarrow
\begin{pmatrix}
4a & 4b & 4c \\
-d & -e & -f \\
g & h & i%
\end{pmatrix}%
=%
\begin{pmatrix}
4a & -b & c \\
4d & -e & f \\
4g & -h & i%
\end{pmatrix}
\\
&\Leftrightarrow &\left\{
\begin{array}{ccc}
4a=4a & & -f=f \\
4b=-b & & g=4g \\
4c=c & & h=-h \\
-d=4d & & i=i \\
-e=-e & &
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{ccc}
0=0 & & -2f=0 \\
5b=0 & & -3g=0 \\
c=0 & & 2h=0 \\
-5d=0 & & 0=0 \\
0=0 & &
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{ccc}
b=0 & & f=0 \\
c=0 & & g=0 \\
d=0 & & h=0%
\end{array}%
\right. \Leftrightarrow X=%
\begin{pmatrix}
a & 0 & 0 \\
0 & e & 0 \\
0 & 0 & i%
\end{pmatrix}%
\end{eqnarray*}
: : :
\item \thinspace
\begin{enumerate}
\item On a $Z^{2}=(Q^{-1}YQ)^{2}=Q^{-1}Y\underset{=I}{\underbrace{QQ^{-1}}}%
YQ=Q^{-1}Y^{2}Q=QM(2,1)Q^{-1}=%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
$ (d'après le rappel de cette partie).\newline
On en déduit que $%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
Z=Z^{2}Z=Z^{3}=ZZ^{2}=Z%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
.$
\item D'après l'égalité $%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
Z=Z%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
,$ on est en droit d'appliquer le résultat de la question 2, ce qui entraine
l'existence de trois réels $\alpha ,\beta ,\gamma $ tels que $Z=%
\begin{pmatrix}
\alpha & 0 & 0 \\
0 & \beta & 0 \\
0 & 0 & \gamma
\end{pmatrix}%
$ et comme $Z^{2}=%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
,$ on en déduit que $%
\begin{pmatrix}
\alpha & 0 & 0 \\
0 & \beta & 0 \\
0 & 0 & \gamma
\end{pmatrix}%
^{2}=%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
$
\item $%
\begin{pmatrix}
\alpha & 0 & 0 \\
0 & \beta & 0 \\
0 & 0 & \gamma
\end{pmatrix}%
^{2}=%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\Leftrightarrow
\begin{pmatrix}
\alpha ^{2} & 0 & 0 \\
0 & \beta ^{2} & 0 \\
0 & 0 & \gamma ^{2}%
\end{pmatrix}%
=%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
\Leftrightarrow \left\{
\begin{array}{c}
\alpha ^{2}=2 \\
\beta ^{2}=-1 \\
\gamma ^{2}=1%
\end{array}%
\right. $\newline
Comme l'équation $\beta ^{2}=-1$ est impossible, on en déduit que la matrice
$Z$ n'existe pas, ce qui entraine l'inexistence de la matrice $Y,$ autrement
dit, la matrice $%
\begin{pmatrix}
4 & 0 & 0 \\
0 & -1 & 0 \\
0 & 0 & 1%
\end{pmatrix}%
$ ne possède pas de racine carrée.
\end{enumerate}
\end{enumerate}
\section*{CORRECTION EXERCICE 1 (EDHEC\ 2005)}
\begin{enumerate}
\item Soient $\alpha ,\beta $ deux réels et $X=\left(
\begin{array}{cc}
a & b \\
c & d%
\end{array}%
\right) $ et $Y=\left(
\begin{array}{cc}
a^{\prime } & b^{\prime } \\
c^{\prime } & d^{\prime }%
\end{array}%
\right) $ deux vecteurs de $\mathfrak{M}_{2}(\mathbb{R}).$ On va calculer $%
f(\alpha X+\beta Y)$ puis on va «séparer la partie en $\alpha $ de la partie
en $\beta $»
\begin{eqnarray*}
f(\alpha X+\beta Y) &=&f(\alpha \left(
\begin{array}{cc}
a & b \\
c & d%
\end{array}%
\right) +\beta \left(
\begin{array}{cc}
a^{\prime } & b^{\prime } \\
c^{\prime } & d^{\prime }%
\end{array}%
\right) )=f(\left(
\begin{array}{cc}
\alpha a+\beta a^{\prime } & \alpha b+\beta b^{\prime } \\
\alpha c+\beta c^{\prime } & \alpha d+\beta d^{\prime }%
\end{array}%
\right) ) \\
&=&\left(
\begin{array}{cc}
\alpha a+\beta a^{\prime } & \alpha b+\beta b^{\prime } \\
\alpha c+\beta c^{\prime } & \alpha d+\beta d^{\prime }%
\end{array}%
\right) +\left[ (\alpha a+\beta a^{\prime })+(\alpha d+\beta d^{\prime })%
\right] \left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) \\
("M &=&\left(
\begin{array}{cc}
\alpha a+\beta a^{\prime } & \alpha b+\beta b^{\prime } \\
\alpha c+\beta c^{\prime } & \alpha d+\beta d^{\prime }%
\end{array}%
\right) ,\quad "a=\alpha a+\beta a^{\prime },\quad d=\alpha d+\beta
d^{\prime }") \\
&=&\alpha \left[ \left(
\begin{array}{cc}
a & b \\
c & d%
\end{array}%
\right) +(a+d)\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) \right] +\beta \left[ \left(
\begin{array}{cc}
a^{\prime } & b^{\prime } \\
c^{\prime } & d^{\prime }%
\end{array}%
\right) +(a^{\prime }+d^{\prime })\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) \right] =\alpha f(X)+\beta f(Y)
\end{eqnarray*}
\item \thinspace
\begin{enumerate}
\item Un calcul direct nous donne%
\begin{eqnarray*}
f(J_{1}) &=&f\left( \left(
\begin{array}{cc}
1 & 0 \\
0 & 0%
\end{array}%
\right) \right) =\left(
\begin{array}{cc}
1 & 0 \\
0 & 0%
\end{array}%
\right) +(1+0)\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =\left(
\begin{array}{cc}
1 & 0 \\
0 & 0%
\end{array}%
\right) +\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =\left(
\begin{array}{cc}
2 & 0 \\
0 & 1%
\end{array}%
\right) =2J_{1}+J_{4} \\
f(J_{2}) &=&f\left( \left(
\begin{array}{cc}
0 & 1 \\
0 & 0%
\end{array}%
\right) \right) =\left(
\begin{array}{cc}
0 & 1 \\
0 & 0%
\end{array}%
\right) +(0+0)\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =\left(
\begin{array}{cc}
0 & 1 \\
0 & 0%
\end{array}%
\right) =J_{2} \\
f(J_{3}) &=&f\left( \left(
\begin{array}{cc}
0 & 0 \\
1 & 0%
\end{array}%
\right) \right) =\left(
\begin{array}{cc}
0 & 0 \\
1 & 0%
\end{array}%
\right) +(0+0)\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =\left(
\begin{array}{cc}
0 & 0 \\
1 & 0%
\end{array}%
\right) =J_{3} \\
f(J_{4}) &=&f\left( \left(
\begin{array}{cc}
0 & 0 \\
0 & 1%
\end{array}%
\right) \right) =\left(
\begin{array}{cc}
0 & 0 \\
0 & 1%
\end{array}%
\right) +(0+1)\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =\left(
\begin{array}{cc}
0 & 0 \\
0 & 1%
\end{array}%
\right) +\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =\left(
\begin{array}{cc}
1 & 0 \\
0 & 2%
\end{array}%
\right) =J_{1}+4J_{4}
\end{eqnarray*}
\item Il s'agit d'exprimer les images de chaque vecteur de la base de
l'espace de départ $(f(J_{1}),f(J_{2}),f(J_{3}),f(J_{4})$ ici) comme
combinaison linéaire des vecteurs de la base de l'espace d'arrivée ($%
J_{1},J_{2},J_{3},J_{4}$ ici). D'après la question précédente, on a
\begin{equation*}
A=\overset{%
\begin{array}{cccc}
f(J_{1}) & f(J_{2}) & f(J_{3}) & f(J_{4})%
\end{array}%
}{%
\begin{pmatrix}
& 2 & & 0 & & 0 & & 1 & \\
& & & & & & & & \\
& 0 & & 1 & & 0 & & 0 & \\
& & & & & & & & \\
& 0 & & 0 & & 1 & & 0 & \\
& & & & & & & & \\
& 1 & & 0 & & 0 & & 2 &
\end{pmatrix}%
}%
\begin{array}{c}
J_{1} \\
\\
J_{2} \\
\\
J_{3} \\
\\
J_{4}%
\end{array}%
\end{equation*}
\end{enumerate}
\item \thinspace
\begin{enumerate}
\item Commençons par montrer qu'il s'agit d'une famille libre. Soient $%
a,b,c,d$ quatre réels tels que
\begin{gather*}
aK_{1}+bK_{2}+cK_{3}+dK_{4}=0_{\mathfrak{M}_{2}(\mathbb{R})}\Leftrightarrow a%
\begin{pmatrix}
1 & 0 \\
0 & -1%
\end{pmatrix}%
+b%
\begin{pmatrix}
0 & 1 \\
0 & 0%
\end{pmatrix}%
+c%
\begin{pmatrix}
0 & 0 \\
1 & 0%
\end{pmatrix}%
+d%
\begin{pmatrix}
1 & 0 \\
0 & 1%
\end{pmatrix}%
=%
\begin{pmatrix}
0 & 0 \\
0 & 0%
\end{pmatrix}
\\
\Leftrightarrow
\begin{pmatrix}
a+d & b \\
c & -a+d%
\end{pmatrix}%
=%
\begin{pmatrix}
0 & 0 \\
0 & 0%
\end{pmatrix}%
\Leftrightarrow \left\{
\begin{array}{c}
a+d=0 \\
b=0 \\
c=0 \\
-a+d=0%
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
a+d=0 \\
b=0 \\
c=0 \\
2d=0%
\end{array}%
\right. \left\vert
\begin{array}{c}
\text{«\emph{Pivot pour }}a\text{»} \\
\\
\\
L_{4}\leftarrow L_{4}+L_{1}%
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
d=0 \\
c=0 \\
b=0 \\
a=0%
\end{array}%
\right.
\end{gather*}
donc la famille $(K_{1},K_{2},K_{3},K_{4})$ est bien libre.\newline
Montrons qu'il s'agit d'une famille génératrice. Soit $X=%
\begin{pmatrix}
x & y \\
z & t%
\end{pmatrix}%
\in \mathfrak{M}_{2}(\mathbb{R})$, existe-t-il quatre réels $a,b,c,d$ tels
que
\begin{eqnarray*}
X &=&aK_{1}+bK_{2}+cK_{3}+dK_{4}\Leftrightarrow
\begin{pmatrix}
x & y \\
z & t%
\end{pmatrix}%
=a%
\begin{pmatrix}
1 & 0 \\
0 & -1%
\end{pmatrix}%
+b%
\begin{pmatrix}
0 & 1 \\
0 & 0%
\end{pmatrix}%
+c%
\begin{pmatrix}
0 & 0 \\
1 & 0%
\end{pmatrix}%
+d%
\begin{pmatrix}
1 & 0 \\
0 & 1%
\end{pmatrix}
\\
&\Leftrightarrow &%
\begin{pmatrix}
x & y \\
z & t%
\end{pmatrix}%
=%
\begin{pmatrix}
a+d & b \\
c & -a+d%
\end{pmatrix}%
\Leftrightarrow \left\{
\begin{array}{c}
a+d=x \\
b=y \\
c=z \\
-a+d=t%
\end{array}%
\right. \Leftrightarrow \left\{
\begin{array}{c}
a+d=x \\
b=y \\
c=z \\
2d=t%
\end{array}%
\right. \left\vert
\begin{array}{c}
\text{«\emph{Pivot pour }}a\text{»} \\
\\
\\
L_{4}\leftarrow L_{4}+L_{1}%
\end{array}%
\right.
\end{eqnarray*}
Par conséquent, on en déduit l'existence de $b,c,$ puis $d$ et enfin celui
de $a$, ce qui entraine que les réels $a,b,c,d$ existent bien. Ainsi, la
famille $(K_{1},K_{2},K_{3},K_{4})$ est génératrice et libre donc c'est une
base de $\mathfrak{M}_{2}(\mathbb{R})$
\item Il s'agit d'exprimer les images de chaque vecteur de la base de
l'espace de départ $(f(K_{1}),f(K_{2}),f(K_{3}),f(K_{4})$ ici) comme
combinaison linéaire des vecteurs de la base de l'espace d'arrivée ($%
K_{1},K_{2},K_{3},K_{4}$ ici). On commence par calculer ces différentes
images%
\begin{eqnarray*}
f(K_{1}) &=&f\left(
\begin{pmatrix}
1 & 0 \\
0 & -1%
\end{pmatrix}%
\right) =%
\begin{pmatrix}
1 & 0 \\
0 & -1%
\end{pmatrix}%
+(1+(-1))\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =%
\begin{pmatrix}
1 & 0 \\
0 & -1%
\end{pmatrix}%
=K_{1} \\
f(K_{2}) &=&f\left(
\begin{pmatrix}
0 & 1 \\
0 & 0%
\end{pmatrix}%
\right) =%
\begin{pmatrix}
0 & 1 \\
0 & 0%
\end{pmatrix}%
+(0+0)\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =%
\begin{pmatrix}
0 & 1 \\
0 & 0%
\end{pmatrix}%
=K_{2} \\
f(K_{3}) &=&f\left(
\begin{pmatrix}
0 & 0 \\
1 & 0%
\end{pmatrix}%
\right) =%
\begin{pmatrix}
0 & 0 \\
1 & 0%
\end{pmatrix}%
+(0+0)\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =%
\begin{pmatrix}
0 & 0 \\
1 & 0%
\end{pmatrix}%
=K_{3} \\
f(K_{4}) &=&f\left(
\begin{pmatrix}
1 & 0 \\
0 & 1%
\end{pmatrix}%
\right) =%
\begin{pmatrix}
1 & 0 \\
0 & 1%
\end{pmatrix}%
+(1+1)\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =%
\begin{pmatrix}
3 & 0 \\
0 & 3%
\end{pmatrix}%
=3\left(
\begin{array}{cc}
1 & 0 \\
0 & 1%
\end{array}%
\right) =3K_{4}
\end{eqnarray*}%
On en déduit immédiatement que%
\begin{equation*}
A=\overset{%
\begin{array}{cccc}
f(K_{1}) & f(K_{2}) & f(K_{3}) & f(K_{4})%
\end{array}%
}{%
\begin{pmatrix}
& 1 & & 0 & & 0 & & 0 & \\
& & & & & & & & \\
& 0 & & 1 & & 0 & & 0 & \\
& & & & & & & & \\
& 0 & & 0 & & 1 & & 0 & \\
& & & & & & & & \\
& 0 & & 0 & & 0 & & 3 &
\end{pmatrix}%
}%
\begin{array}{c}
K_{1} \\
\\
K_{2} \\
\\
K_{3} \\
\\
K_{4}%
\end{array}%
\end{equation*}
\end{enumerate}
\end{enumerate}
\section*{CORRECTION\ EXERCICE\ 2 (EM LYON 1998)}
\begin{enumerate}
\item \thinspace
\begin{enumerate}
\item La fonction $t\mapsto \dfrac{1}{t^{4}+1}$ est continue sur $\mathbb{R}%
=]-\infty ,+\infty \lbrack $ (quotient de fonctions continues sur $\mathbb{R}
$ dont le dénominateur ne s'annule pas sur $\mathbb{R})$ et $0\in ]-\infty
,+\infty \lbrack $ donc la fonction $x\mapsto \dint\limits_{0}^{x}\dfrac{1}{%
t^{4}+1}dt=G(x)$ est définie et $C^{1}$ sur $\mathbb{R}$ et $\forall x\in
\mathbb{R},\quad G^{\prime }(x)=\dfrac{1}{x^{4}+1}$. Par conséquent, la
fonction $G$ est définie sur $\mathbb{R}.$
\item On l'a répondu à cette question à la question 1.a)
\end{enumerate}
\item \thinspace
\begin{enumerate}
\item La fonction $t\mapsto \dfrac{1}{t^{4}+1}$ est continue sur $\mathbb{R}%
=]-\infty ,+\infty \lbrack $. On doit exiger que $x$ et $2x$ appartiennent à$%
\mathbb{\ }]-\infty ,+\infty \lbrack $ (au même intervalle, mais il n'y en a
qu'un ici), ce qui est toujours vraie, donc l'intégrale $\dint%
\limits_{x}^{2x}\dfrac{1}{t^{4}+1}dt$ existe pour tout réel $x.$
\item \textbf{Premier cas} $2x\geqslant x\Leftrightarrow x\geqslant 0$
(bornes dans l'ordre croissant). La fonction $t\mapsto \dfrac{1}{t^{4}+1}$
est positive $\mathbb{R}_{+}$ donc sur $[x,2x]\subset \mathbb{R}_{+}$. En
outre, l'intégration portant sur des bornes croissantes $(2x\geqslant x)$,
on en déduit que l'intégrale $\dint\limits_{x}^{2x}\dfrac{1}{t^{4}+1}dt=F(x)$
est positive\newline
\textbf{Second cas} $2x\leqslant x\Leftrightarrow x\leqslant 0$ (bornes dans
l'ordre décroissant). On commence par utiliser la relation de Chasles%
\begin{equation*}
F(x)=\dint\limits_{x}^{2x}\dfrac{1}{t^{4}+1}dt=-\dint\limits_{2x}^{x}\dfrac{1%
}{t^{4}+1}dt
\end{equation*}%
La fonction $t\mapsto \dfrac{1}{t^{4}+1}$ est positive sur $\mathbb{R}_{-}$
donc sur $[2x,x]\subset \mathbb{R}_{-}.$ En outre, l'intégration dans l'inté%
grale $\dint\limits_{2x}^{x}\dfrac{1}{t^{4}+1}dt$ portant sur des bornes
croissantes $(2x\leqslant x),$ on en déduit que l'intégrale $%
\dint\limits_{2x}^{x}\dfrac{1}{t^{4}+1}dt$ est positive, ce qui implique que
$F(x)=-\dint\limits_{2x}^{x}\dfrac{1}{t^{4}+1}dt$ est négative.\newline
Par conséquent, la fonction $F$ est négative sur $\mathbb{R}_{-}$ et
positive sur $\mathbb{R}_{+}$.
\item Pour commencer, l'ensemble de définition de $F,$ qui est $\mathbb{R},$
est symétrique par rapport à $0.$ Ensuite, en utilisant le changement de
variable $u=-t$ dans l'intégrale définissant $F(-x),$ on obtient
\begin{equation*}
F(-x)=\dint\limits_{(-x)}^{2(-x)}\dfrac{1}{t^{4}+1}dt=\dint\limits_{x}^{2x}%
\dfrac{1}{(-u)^{4}+1}(-du)=-\dint\limits_{x}^{2x}\dfrac{1}{u^{4}+1}du=-F(x)
\end{equation*}%
donc la fonction $F$ est impaire.
\end{enumerate}
\item \thinspace
\begin{enumerate}
\item On encadre "à la main" la fonction $t\mapsto \dfrac{1}{t^{4}+1}$ sur
l'intervalle $[x,2x]$ lorsque $x>0.$%
\begin{eqnarray*}
x &\leqslant &t\leqslant 2x\Rightarrow x^{4}\leqslant t^{4}\leqslant
16x^{4}\Rightarrow x^{4}+1\leqslant t^{4}+1\leqslant 16x^{4}+1 \\
&\Rightarrow &\dfrac{1}{16x^{4}+1}\leqslant \dfrac{1}{t^{4}+1}\leqslant
\dfrac{1}{x^{4}+1}\qquad \forall t\in \lbrack x,2x]
\end{eqnarray*}%
Puisque $x\leqslant 2x$ (car $x>0),$ en intégrant l'inégalité précédente sur
$[x,2x]$ (l'intégration portant sur des bornes croissantes), on obtient%
\begin{eqnarray*}
\dint\limits_{x}^{2x}\dfrac{dt}{16x^{4}+1} &\leqslant &\dint\limits_{x}^{2x}%
\dfrac{dt}{t^{4}+1}\leqslant \dint\limits_{x}^{2x}\dfrac{dt}{x^{4}+1}%
\Leftrightarrow \dfrac{1}{16x^{4}+1}\dint\limits_{x}^{2x}1dt\leqslant
F(x)\leqslant \dfrac{1}{x^{4}+1}\dint\limits_{x}^{2x}1dt \\
&\Leftrightarrow &\dfrac{1}{16x^{4}+1}\left[ t\right] _{t=x}^{t=2x}dt%
\leqslant F(x)\leqslant \dfrac{1}{x^{4}+1}\left[ t\right] _{t=x}^{t=2x}%
\Leftrightarrow \dfrac{x}{16x^{4}+1}\leqslant F(x)\leqslant \dfrac{x}{x^{4}+1%
}
\end{eqnarray*}
\item Puisque l'on a l'encadrement $\forall x\geqslant 0,\quad \dfrac{x}{%
16x^{4}+1}\leqslant F(x)\leqslant \dfrac{x}{x^{4}+1}$ et
\begin{eqnarray*}
\dfrac{x}{16x^{4}+1}\underset{x\rightarrow +\infty }{\sim }\dfrac{x}{16x^{4}}
&=&\dfrac{1}{16x^{3}}\underset{x\rightarrow +\infty }{\rightarrow }%
0\Rightarrow \lim\limits_{x\rightarrow +\infty }\dfrac{x}{16x^{4}+1}=0 \\
\dfrac{x}{x^{4}+1}\underset{x\rightarrow +\infty }{\sim }\dfrac{x}{x^{4}} &=&%
\dfrac{1}{x^{3}}\underset{x\rightarrow +\infty }{\rightarrow }0\Rightarrow
\lim\limits_{x\rightarrow +\infty }\dfrac{x}{x^{4}+1}=0,
\end{eqnarray*}%
le théorème d'encadrement montre que $\lim\limits_{x\rightarrow +\infty
}F(x)=0.$ Par imparité, on en déduit que $\lim\limits_{x\rightarrow -\infty
}F(x)=-0=0.$
\item On commence par remarquer que%
\begin{equation*}
\forall x\in \mathbb{R},\quad F(x)=\dint\limits_{x}^{2x}\dfrac{dt}{t^{4}+1}%
=\dint\limits_{0}^{2x}\dfrac{dt}{t^{4}+1}-\dint\limits_{0}^{x}\dfrac{dt}{%
t^{4}+1}=G(2x)-G(x)
\end{equation*}%
La fonction $G$ étant dérivable sur $\mathbb{R}$ ainsi que la fonction $%
x\mapsto 2x,$ on en déduit que la fonction $F$ est dérivable sur $\mathbb{R}$
tout entier et
\begin{equation*}
\forall x\in \mathbb{R},\quad F^{\prime }(x)=(G(2x))^{\prime
}-(G(x))^{\prime }=(2x)^{\prime }G^{\prime }(2x)-G^{\prime }(x)=2\times
\dfrac{1}{(2x)^{4}+1}-\dfrac{1}{x^{4}+1}=\dfrac{2}{16x^{4}+1}-\dfrac{1}{%
x^{4}+1}
\end{equation*}
\end{enumerate}
\end{enumerate}
\label{fin}
\end{document}