[求助]随机数的概率密度函数
现有一组零均值得随机数,如何求得其概率密度函数re:
(1) 统计随机数总数total<BR>(2)将-inf +inf 画分成若干区间,足够细的<BR>(3)统计在每一个区间的随机数个数,比如i个区间Ni<BR>(4)密度就是Ni/Total<BR> 按照上述方法求得的应该是概率分布函数吧<BR>我想问一下怎么样才能得到概率密度函数 以上就是概率密度函数,不是概率分布函数<BR> <P>我算出来结果好象不对啊<BR>似乎应该是Ni/(N*W)W为区间宽度,但也不太对</P> <P>for example: the generated random number is <br><br> 0,0,0,1,-1,-1,1,1,<br><br>the amount of 0 is 3<br>the amount of 1 is 3 <br>the amount of -1 is 2 <br>the total amount is 8,so the density is 3/8,3/8,2/8 respectively. As is the <br>same case for continous example. You know the density f should satisfy<br> integrationfdx=1 .....continous example<br> sum f=1 ....scatter example <br>3/8+3/8+2/8=1 </P>[此贴子已经被作者于2006-5-25 14:38:12编辑过]
P=p*dx,所以现在要求的是p而不是P=p*dx。上面的例子求出来的应该是P吧
页:
[1]