[問題] 如何證明i.i.d.
各位好，小弟在此有個小問題請教，假設一個由randn產生的矩陣如下：
A = randn(2,2,100)+j*randn(2,2,100);
B1 = A(1,1,:);
B2 = A(1,2,:);
B3 = A(2,1,:);
B4 = A(2,2,:);
那請問如何證明B1~B4互相都是i.i.d.呢？
是用xcorr 去求corss-correlation嗎？
假若B1~B4並非為i.i.d.那如何去產生一個都為i.i.d.且mean = 0,var = sigma^2
的A矩陣呢？
以上，麻煩各位解惑 ^^
--
※ 發信站: 批踢踢實業坊(ptt.cc)
◆ From: 140.116.187.145
※ jack555:轉錄至看板 MATLAB 11/24 14:33
推
11/24 15:01, , 1F
11/24 15:01, 1F
→
11/24 15:02, , 2F
11/24 15:02, 2F
→
11/24 15:03, , 3F
11/24 15:03, 3F
→
11/24 15:24, , 4F
11/24 15:24, 4F
→
11/24 15:25, , 5F
11/24 15:25, 5F
→
11/24 15:39, , 6F
11/24 15:39, 6F
→
11/24 15:40, , 7F
11/24 15:40, 7F
→
11/24 15:56, , 8F
11/24 15:56, 8F
推
11/24 15:59, , 9F
11/24 15:59, 9F
→
11/24 16:00, , 10F
11/24 16:00, 10F
→
11/24 16:05, , 11F
11/24 16:05, 11F
→
11/24 16:05, , 12F
11/24 16:05, 12F
→
11/24 16:08, , 13F
11/24 16:08, 13F
→
11/24 16:08, , 14F
11/24 16:08, 14F
→
01/02 15:00,
7年前
, 15F
01/02 15:00, 15F