首页 > 代码库 > C++与正态分布
C++与正态分布
从上图可以看出,当相差1个方差(σ), 满足要求的面积有68.27%.
当相差2个方差(σ)时,满足要求的面积有95.45.
当相差3个方差(σ)时,满足要求的面积有99.73%.
满足标准正态分的曲线,可以查表来求得正态分布的幅度.(见文后所附表格)
方差(Variance),是各个数据分别与其和的平均数之差的平方的和的平均数,用字母D表示。在概率论和数理统计中,方差用来度量随机变量和其数学期望(即均值)之间的偏离程度。
标准差(StandardDeviation),是离均差平方和平均后的方根,用σ表示。标准差是方差的算术平方根。标准差能反映一个数据集的离散程度。
测试代码:
// NormalDistribution.cpp : Defines the entry point for the console application. // #include <stdio.h> #include <tchar.h> #include <iostream> #include <windows.h> #include <algorithm> #define _USE_MATH_DEFINES #include <math.h> using namespace std; // 高斯分布随机数系列,默认期望值为0,方差为1 double GaussRand(double dExpect = 0, double dVariance = 1); double GaussRand(double dExpect, double dVariance) { static double V1, V2, S; static int phase = 0; double X; if ( phase == 0 ) { do { double U1 = (double)rand() / RAND_MAX; double U2 = (double)rand() / RAND_MAX; V1 = 2 * U1 - 1; V2 = 2 * U2 - 1; S = V1 * V1 + V2 * V2; } while(S >= 1 || S == 0); X = V1 * sqrt(-2 * log(S) / S); } else { X = V2 * sqrt(-2 * log(S) / S); } phase = 1 - phase; return (X*dVariance + dExpect); } int _tmain(int argc, _TCHAR* argv[]) { const int DATA_CNT = 100000; double dArrData[DATA_CNT] = {0}; double dSum = 0; // 对所有数赋随机数,默认期望值为0,方差为1 srand(GetTickCount()); for (int nIdx = 0; nIdx < DATA_CNT; nIdx++) { // 防止计算方差时数值过大 dArrData[nIdx] = GaussRand(); dSum += dArrData[nIdx]; } // 求平均数 double dAverageData = http://www.mamicode.com/dSum / DATA_CNT; >
(附)标准正态分布表
φ( - x ) = 1 –φ( x )
x | 0 | 0.01 | 0.02 | 0.03 | 0.04 | 0.05 | 0.06 | 0.07 | 0.08 | 0.09 |
0 | 0.500 0 | 0.504 0 | 0.508 0 | 0.512 0 | 0.516 0 | 0.519 9 | 0.523 9 | 0.527 9 | 0.531 9 | 0.535 9 |
0.1 | 0.539 8 | 0.543 8 | 0.547 8 | 0.551 7 | 0.555 7 | 0.559 6 | 0.563 6 | 0.567 5 | 0.571 4 | 0.575 3 |
0.2 | 0.579 3 | 0.583 2 | 0.587 1 | 0.591 0 | 0.594 8 | 0.598 7 | 0.602 6 | 0.606 4 | 0.610 3 | 0.614 1 |
0.3 | 0.617 9 | 0.621 7 | 0.625 5 | 0.629 3 | 0.633 1 | 0.636 8 | 0.640 4 | 0.644 3 | 0.648 0 | 0.651 7 |
0.4 | 0.655 4 | 0.659 1 | 0.662 8 | 0.666 4 | 0.670 0 | 0.673 6 | 0.677 2 | 0.680 8 | 0.684 4 | 0.687 9 |
0.5 | 0.691 5 | 0.695 0 | 0.698 5 | 0.701 9 | 0.705 4 | 0.708 8 | 0.712 3 | 0.715 7 | 0.719 0 | 0.722 4 |
0.6 | 0.725 7 | 0.729 1 | 0.732 4 | 0.735 7 | 0.738 9 | 0.742 2 | 0.745 4 | 0.748 6 | 0.751 7 | 0.754 9 |
0.7 | 0.758 0 | 0.761 1 | 0.764 2 | 0.767 3 | 0.770 3 | 0.773 4 | 0.776 4 | 0.779 4 | 0.782 3 | 0.785 2 |
0.8 | 0.788 1 | 0.791 0 | 0.793 9 | 0.796 7 | 0.799 5 | 0.802 3 | 0.805 1 | 0.807 8 | 0.810 6 | 0.813 3 |
0.9 | 0.815 9 | 0.818 6 | 0.821 2 | 0.823 8 | 0.826 4 | 0.828 9 | 0.835 5 | 0.834 0 | 0.836 5 | 0.838 9 |
1 | 0.841 3 | 0.843 8 | 0.846 1 | 0.848 5 | 0.850 8 | 0.853 1 | 0.855 4 | 0.857 7 | 0.859 9 | 0.862 1 |
1.1 | 0.864 3 | 0.866 5 | 0.868 6 | 0.870 8 | 0.872 9 | 0.874 9 | 0.877 0 | 0.879 0 | 0.881 0 | 0.883 0 |
1.2 | 0.884 9 | 0.886 9 | 0.888 8 | 0.890 7 | 0.892 5 | 0.894 4 | 0.896 2 | 0.898 0 | 0.899 7 | 0.901 5 |
1.3 | 0.903 2 | 0.904 9 | 0.906 6 | 0.908 2 | 0.909 9 | 0.911 5 | 0.913 1 | 0.914 7 | 0.916 2 | 0.917 7 |
1.4 | 0.919 2 | 0.920 7 | 0.922 2 | 0.923 6 | 0.925 1 | 0.926 5 | 0.927 9 | 0.929 2 | 0.930 6 | 0.931 9 |
1.5 | 0.933 2 | 0.934 5 | 0.935 7 | 0.937 0 | 0.938 2 | 0.939 4 | 0.940 6 | 0.941 8 | 0.943 0 | 0.944 1 |
1.6 | 0.945 2 | 0.946 3 | 0.947 4 | 0.948 4 | 0.949 5 | 0.950 5 | 0.951 5 | 0.952 5 | 0.953 5 | 0.953 5 |
1.7 | 0.955 4 | 0.956 4 | 0.957 3 | 0.958 2 | 0.959 1 | 0.959 9 | 0.960 8 | 0.961 6 | 0.962 5 | 0.963 3 |
1.8 | 0.964 1 | 0.964 8 | 0.965 6 | 0.966 4 | 0.967 2 | 0.967 8 | 0.968 6 | 0.969 3 | 0.970 0 | 0.970 6 |
1.9 | 0.971 3 | 0.971 9 | 0.972 6 | 0.973 2 | 0.973 8 | 0.974 4 | 0.975 0 | 0.975 6 | 0.976 2 | 0.976 7 |
2 | 0.977 2 | 0.977 8 | 0.978 3 | 0.978 8 | 0.979 3 | 0.979 8 | 0.980 3 | 0.980 8 | 0.981 2 | 0.981 7 |
2.1 | 0.982 1 | 0.982 6 | 0.983 0 | 0.983 4 | 0.983 8 | 0.984 2 | 0.984 6 | 0.985 0 | 0.985 4 | 0.985 7 |
2.2 | 0.986 1 | 0.986 4 | 0.986 8 | 0.987 1 | 0.987 4 | 0.987 8 | 0.988 1 | 0.988 4 | 0.988 7 | 0.989 0 |
2.3 | 0.989 3 | 0.989 6 | 0.989 8 | 0.990 1 | 0.990 4 | 0.990 6 | 0.990 9 | 0.991 1 | 0.991 3 | 0.991 6 |
2.4 | 0.991 8 | 0.992 0 | 0.992 2 | 0.992 5 | 0.992 7 | 0.992 9 | 0.993 1 | 0.993 2 | 0.993 4 | 0.993 6 |
2.5 | 0.993 8 | 0.994 0 | 0.994 1 | 0.994 3 | 0.994 5 | 0.994 6 | 0.994 8 | 0.994 9 | 0.995 1 | 0.995 2 |
2.6 | 0.995 3 | 0.995 5 | 0.995 6 | 0.995 7 | 0.995 9 | 0.996 0 | 0.996 1 | 0.996 2 | 0.996 3 | 0.996 4 |
2.7 | 0.996 5 | 0.996 6 | 0.996 7 | 0.996 8 | 0.996 9 | 0.997 0 | 0.997 1 | 0.997 2 | 0.997 3 | 0.997 4 |
2.8 | 0.997 4 | 0.997 5 | 0.997 6 | 0.997 7 | 0.997 7 | 0.997 8 | 0.997 9 | 0.997 9 | 0.998 0 | 0.998 1 |
2.9 | 0.998 1 | 0.998 2 | 0.998 2 | 0.998 3 | 0.998 4 | 0.998 4 | 0.998 5 | 0.998 5 | 0.998 6 | 0.998 6 |
x | 0 | 0.1 | 0.2 | 0.3 | 0.4 | 0.5 | 0.6 | 0.7 | 0.8 | 0.9 |
3 | 0.998 7 | 0.999 0 | 0.999 3 | 0.999 5 | 0.999 7 | 0.999 8 | 0.999 8 | 0.999 9 | 0.999 9 | 1.000 0 |
(附)正态分布概率表
Φ( u ) =
C++与正态分布