numpy笔记案例之统计运算

一、统计运算

1.统计指标函数
min，max，mean，median，var，std
np.函数名（数组，axis=1）求每一行的
ndarray.方法（axis=0）求每一列的
0为列，1为行，（数字为维度，如shape=（4，4）的时候0表示第一个维度，1第二个维度（-1）也可）
tips:不同的API，axis轴可能也不同，根据实际情况查询
随机生成八只股票两周的交易日涨幅数据

>>> import numpy as np
>>> data=np.random.normal(loc=0,scale=1,size=(8,10))
>>> data
array([[-0.82819181, -0.94524632,  1.00261548,  1.22347942,  3.1216305 ,
         1.26492343, -0.43991794,  1.15560652, -1.14764635, -0.10755365],
       [ 0.10702001,  0.32275393,  0.36884494, -0.80633473,  0.10975151,
         0.26445976,  2.51025557, -0.34614923,  1.61700407,  0.92735698],
       [-1.52444973,  0.37116937,  0.20480437,  0.45760614,  0.58600911,
        -0.03873739,  0.05557741, -0.37362175,  0.21372127, -0.21417988],
       [ 0.04557212, -0.60571389, -0.4828674 , -1.3691487 , -0.78917793,
         0.77271334,  0.53532935, -0.82460714,  0.14611143, -1.73388128],
       [-1.39373237,  0.66582896,  0.20633859,  0.27751827, -1.21031274,
         0.33675267,  0.16496407, -1.20571003, -0.63252079,  2.25212589],
       [-0.7853435 ,  0.45017614,  0.88711645, -0.57357384,  1.27036775,
        -1.42000162,  1.47493257,  0.32056218, -0.0735568 ,  0.49032614],
       [-0.17759054, -1.17023642, -2.01238529,  1.74187624, -0.74469437,
         1.19414198, -1.77822096, -0.40940959,  1.42971513,  1.58769954],
       [ 0.37682035, -0.28473632,  1.66208163,  0.95411509, -0.64629404,
        -2.09255195,  0.90873249,  0.27206528, -0.05181042,  1.96627027]])
>>> temp=data[0:4,0:4]
>>> temp
array([[-0.82819181, -0.94524632,  1.00261548,  1.22347942],
       [ 0.10702001,  0.32275393,  0.36884494, -0.80633473],
       [-1.52444973,  0.37116937,  0.20480437,  0.45760614],
       [ 0.04557212, -0.60571389, -0.4828674 , -1.3691487 ]])
>>> np.max(temp)#最大值
1.2234794157101396
>>> np.max(temp,axis=0)
array([0.10702001, 0.37116937, 1.00261548, 1.22347942])#每一列的最大值
>>> temp.max()
1.2234794157101396
>>> temp.max(axis=1)
array([1.22347942, 0.36884494, 0.45760614, 0.04557212])#每一行的最大值
>>> np.min(temp)#最小值
-1.524449725633864
>>> temp.min()#以下ndarray.方法同理
-1.524449725633864 
>>> np.mean(temp)#平均值
-0.15363042521833595
>>> np.median(temp)#中位数
0.07629606325878743
>>> np.var(temp)#方差
0.6122186110576804
>>> np.std(temp)#标准差
0.7824439986719052

统计某一支股票在某个交易日涨幅最大或者最小
（最大最小值出现的位置）

>>> np.argmax(temp,axis=1)
array([3, 2, 3, 0], dtype=int64)
>>> np.argmin(temp,axis=1)
array([1, 3, 0, 3], dtype=int64)