일부 통신원이 자기 상관 계산 방법에 관해 흥미로운 의문을 제기했기 때문에 시계열과 자기 상관에 대한 지식이 거의없이이를 다루기 시작했습니다.
통신원은 자신의 데이터를 정리했습니다 ( 시계열의 데이터 포인트)를 제외하고 각각 한 시간 지연 씩 이동하여 첫 번째 행이 원래 데이터이고 두 번째 행이 데이터를 이동 한 데이터 (내가 이해 한 것처럼) 시간 단위, 다른 행에 의한 다음 행 등. 끝 부분을 꼬리에 붙임으로써 "원형"데이터 세트를 만들어서이를 추가로 실현했습니다.
그런 다음 무엇이 나올지 알아보기 위해 상관 행렬을 계산하고이 주요 구성 요소를 사용했습니다. 놀랍게도 나는 주파수 분해의 이미지를 얻었고 (다른 데이터와 함께) 하나의 주파수를 보았습니다. 데이터는 첫 번째 주요 구성 요소에 있었고 두 번째 기간에는 두 번째 PC에 있었으며 고유 값을 가진 "관련된"PC ). 먼저 이것이 입력 데이터에 달려 있다고 생각했지만 이제는 순환 이동 ( "Toeplitz"매트릭스라고도 함)을 사용하여 데이터 세트를 특수하게 구성하여 체계적으로이 방식으로 가정합니다. varimax 또는 다른 회전 기준에 대한 PC- 솔루션의 회전은 약간 다른, 아마도 흥미로운 결과를 제공했지만 일반적으로 그러한 주파수 분해를 제공하는 것으로 보입니다.
여기에 대한 링크입니다 내가 만든 한 사진 으로부터는포인트 데이터 세트; 곡선은 단순히 factormatrix의 하중으로 만들어집니다. 하나는 하나의 요인에 대한 하중입니다. 첫 번째 PC1의 곡선은 가장 높은 진폭을 보여 주어야합니다 (대략 최대 적재량의 제곱을 갖기 때문에)
질문 :
- Q1 : 의도적으로 설계된 기능입니까? (이 유형의 데이터 세트를 가진 PCA의)
- Q2 : 주파수 / 파장 분석에 대한 진지한 접근에이 방법이 실제로 사용 가능한가?
[업데이트] 여기에 데이터 세트가 있습니다 (복사 가능함).
-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4
-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5
-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3
0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1
2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0
4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2
6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4
5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6
3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5
1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3
1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1
0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1
-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0
-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2
-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3
0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1
3,5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0
5,7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3
7,6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5
6,7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7
7,5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6
5,4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7
4,3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5
3,2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4
2,3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3
3,5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2
5,4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3
4,3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5
3,2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4
2,3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3
3,4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2
4,-5,-3,-1,0,2,4,6,5,3,1,1,0,-2,-3,-1,0,3,5,7,6,7,5,4,3,2,3,5,4,3,2,3