Math is a big interest for me. I This blog will contain proof of interesting mathematics problems which covers Geometry, Algebra, Number theory and Combinatorics with various solution and generalizations.And I will try to make each post enjoyable to read and useful to the reader. (By my bad English writing, some problem may be post in Chinese.)
Translate
Wednesday, September 25, 2013
|
|
The product of 3 consecutive positive integers is never a square. |
Proof: if we have $a(a^2-1)=b^2$, then$\gcd(a,a^2-1)=\gcd(a^2-1,a^2)=1$, so$a=m^2, a^2-1=n^2$, where$mn=b$, and$\gcd(m,n)=1$. But $a^2-1=n^2$ has no integer solution.
Wednesday, September 18, 2013
|
|
\sqrt[n]{n} |
Let $n$ be an positive integer, prove that \[\sqrt{n}\leqslant\sqrt[n]{n!}\leqslant\frac{n+1}{2}\]
|
|
Inequality from "Math Bar" |
设 $a_{i}>0$
求证:\[\sum_{i=1}^{n}\left(\frac{a_{i}}{a_{i+1}}\right)^{n}\ge\sum_{i=1}^{n}\frac{a_{i+1}}{a_{i}}\]
令$x_{i}=\frac{a_{i}}{a_{i+1}}$
此题可考虑 $(x-1)^2(x^n+x^{n-1}+...+x+1)\ge 0$ 得到$x^{n}\ge x^{n-1}+x-1$再用排序和均值不等式即可。
想复杂了,直接均值也可以,以后做题不能硬套一个思路能简单就尽量简单得好
|
|
The last digital of n! is never periodic (Chinese version) |
证明:假设是周期数列。设周期长度为$T$($T$不一定是最小的)。由抽屉原理,在\[10^u, 10^{u+1},...,10^{u+t-1}\]
中必有两个数模$T$同余。 设这两个数是$10^k$ 与 $10^{k+r}$.于是$T|10^{k+r}-10^k$.
考虑到$2\times 10^{k+r}-10^k$的尾数是$9$.
并\[(2\times 10^{k+r}-10^k)!=(2\times 10^{k+r}-10^k-1)!(2\times 10^{k+r}-10^k)\],
等式左边的尾数是$(10^k)!$的尾数, 等式右边则是$(10^k-1)!\times9$的尾数,
而$(10^k)!$的尾数与$(10^k-1)!$相同,于是$(10^k)!$的尾数只能是$5$.但$(10^k)!$显然是个偶数,矛盾。 证毕。
中必有两个数模$T$同余。 设这两个数是$10^k$ 与 $10^{k+r}$.于是$T|10^{k+r}-10^k$.
考虑到$2\times 10^{k+r}-10^k$的尾数是$9$.
并\[(2\times 10^{k+r}-10^k)!=(2\times 10^{k+r}-10^k-1)!(2\times 10^{k+r}-10^k)\],
等式左边的尾数是$(10^k)!$的尾数, 等式右边则是$(10^k-1)!\times9$的尾数,
而$(10^k)!$的尾数与$(10^k-1)!$相同,于是$(10^k)!$的尾数只能是$5$.但$(10^k)!$显然是个偶数,矛盾。 证毕。
|
|
Strategy to find the sum |
\[\sum_{k=45^{\circ}}^{133^{\circ}}\frac{1}{\sin(k)\sin(k+1^{\circ})}\]
part of solution
let\[\sin(x)=\frac{e^{ix}-\frac{1}{e^{ix}}}{2}\]
then \[\frac{1}{\sin(k)\sin(k+1^{\circ})}=\frac{4e^{(2k+1)i}}{(e^{2ki}-1)(e^{(2k+2)i}-1)}\]\[=\left(\frac{1}{e^{2ki}-1}-\frac{1}{e^{(2k+2)i}-1}\right)\cdot\frac{1}{e^i(e^{2i}-1)}\]
Monday, September 16, 2013
|
|
Max and Min |
Let $x,y>0$. Show that \[\max\{\min\{x,y,\frac{1}{x}+\frac{1}{y}\}\}=\min\{\max\{x,y,\frac{1}{x}+\frac{1}{y}\}\}=\sqrt{2}\]
Proof:
\[\left(\max\{x,y,\frac{1}{x}+\frac{1}{y}\}\right)^2\geqslant\frac{(x+y)(\frac{1}{x}+\frac{1}{y})}{2}\geqslant2\]
Surpose that $y\geqslant x$, we have \[\left(\min\{x,y,\frac{1}{x}+\frac{1}{y}\}\right)^2\leqslant x\left(\frac{1}{x}+\frac{1}{y}\right)=1+\frac{x}{y}\leqslant2\]
Then it shows that $\max=\sqrt{2}$
Proof:
\[\left(\max\{x,y,\frac{1}{x}+\frac{1}{y}\}\right)^2\geqslant\frac{(x+y)(\frac{1}{x}+\frac{1}{y})}{2}\geqslant2\]
Surpose that $y\geqslant x$, we have \[\left(\min\{x,y,\frac{1}{x}+\frac{1}{y}\}\right)^2\leqslant x\left(\frac{1}{x}+\frac{1}{y}\right)=1+\frac{x}{y}\leqslant2\]
Then it shows that $\max=\sqrt{2}$
Friday, September 13, 2013
|
|
Cauchy |
若正数$x,y,z$满足$x+y+z=1$,求证:\[\frac{x^4}{z(1-z^2)}+\frac{y^4}{x(1-x^2)}+\frac{z^4}{y(1-y^2)}\geqslant\frac{1}{8}\]
证明:由柯西不等式有\[\frac{x^4}{z(1-z^2)}+\frac{y^4}{x(1-x^2)}+\frac{z^4}{y(1-y^2)}\geqslant\frac{(x^2+y^2+z^2)^2}{x+y+z-x^3-y^3-z^3}=\]
\[\frac{(x^2+y^2+z^2)^2}{1-(x+y+z)(x^3+y^3+z^3)}\geqslant\frac{1/9}{1-1/9}=\frac{1}{8}\]
证明:由柯西不等式有\[\frac{x^4}{z(1-z^2)}+\frac{y^4}{x(1-x^2)}+\frac{z^4}{y(1-y^2)}\geqslant\frac{(x^2+y^2+z^2)^2}{x+y+z-x^3-y^3-z^3}=\]
\[\frac{(x^2+y^2+z^2)^2}{1-(x+y+z)(x^3+y^3+z^3)}\geqslant\frac{1/9}{1-1/9}=\frac{1}{8}\]
|
|
Polynomial |
Let $a,b,c$ be 3 distinct integers, $P(x)$ is a polynomial with integer coefficient, prove that we cannot have $P(a)=b,P(b)=c,$ and $P(c)=a$(USAMO1974)
Prove. we assume that there exsits a $P(x)$ satisfied the conditions. Notice that $a-b|P(a)-P(b)=b-c$ by the same method, we have
$a-b|b-c,b-c|c-a,c-a|a-b$ so$|a-b|=|b-c|=|c-a|$, we only need to consider the situation without abs(because other case are obviously contradiction) $a-b\le b-c,b-c\le c-a, c-a\le a-b$, but $a-b+b-c+c-a=0$, therefore we have $a-b=b-c=c-a$
let $b=c+d,a=c+2d$,we have$c-a=-2d=d$ hence $d=0\iff a=b=c $ which is a contradiction.
Subscribe to:
Comments (Atom)