하루y

### 캔들
 - 빨간색, 흰색 캔들: 상승
 - 파란색, 검정 캔들: 하락

1. 한 개의 캔들
1.1 상승샅바형캔들, Bullish Belt Hold Line (↑71%, ↓29%, +0.42)
1.2 하락샅바형캔들, Bearish Belt Hold Line (↑32%, ↓68%, -0.36)
1.3 망칭형, Hammer (↑60%, ↓40%, +0.2)
1.4 역망치 캔들, Inverted Hammer (↑35%, ↓65%, -0.3)
1.5 호리병 캔들, Takuri Line (↑66%, ↓34%, +0.32)

2. 두 개의 캔들
2.1 하락장악형 패턴, Bear Engulfing (↑21%, ↓79%, -0.58)
2.2 상승장악형 패턴, Bullish Engulfind (↑63%, ↓37%, +0.26)
2.3 상승지속형 패턴, Last Engulfing Top (↑68%, ↓32%, +0.36)
2.4 하락지속형 패턴, Last Engulfind Bottom (↑35%, ↓65%, -0.3)
2.5 상승형 십자캔들 별 패턴, Bullish Doji Star (↑36%, ↓64%, -0.28)
2.6 하락 십자캔들 별 패턴, Bearish Doji Star (↑69%, ↓31%, +0.38)
2.7 관통형 패턴, Piercing Line (↑64%, ↓36%, +0.28)
2.8 먹구름형패턴, Dark Cloud Cover (↑40%, ↓60%, -0.2)
2.9 중심선 위 상승 패턴, Above the Stomach (↑44%, ↓66%, -0.22)
2.10 중심선 아래 하락 패턴, Below the Stomach (↑40%, ↓60%, -0.2)
2.11 상승형 격자 패턴, Windows Rising (↑75%, ↓25%, +0.5)
2.12 하락형 격자 패턴, Windows Falling (↑33%, ↓67%, -0.34)
2.13 상승형 이별 패턴, Bullish Separating Lines (↑72%, ↓28%, +0.44)
2.14 하락형 이별 패턴, Bearish Separating Lines (↑37%, ↓63%, -0.26)

3. 세 개의 캔들
3.1 새벽별 패턴, Moring Star (↑78%, ↓22%, +0.56)
3.2 저녁별 패턴, Evening Star (↑28%, ↓72%, -0.44)
3.3 상승고아형, Bullish Abandoned Baby (↑70%, ↓30%, +0.4)
3.4 하락고아형, Bearish Abandoned Baby (↑31%, ↓69%, -0.38)
3.5 상승 막대샌드위치형 패턴, Bullish Stick Sandwich (↑38%, ↓62%, -0.24)
3.6 하락 막대샌드위치형 패턴, Bearish Stick Sandwich (↑62%, ↓38%, +0.24)
3.7 바닥 삼성형, Three Stars in the South (↑86%, ↓14%, +0.72)
3.8 천장 삼성형, Three Stars in the North (↑14%, ↓86%, -0.72)
3.9 적상병, Three White Soldiers (↑82%, ↓18%, +0.64)
3.10 흑삼병, Three Black Crows (↑22%, ↓78%, -0.56)
3.11 동일흑삼병, Identical Three Crows (↑21%, ↓79%, -0.58)
3.12 상승블록형, Advance Block (↑64%, ↓36%, +0.28)
3.13 지연형, Deliberation-Stalled Pattern (↑77%, ↓23%, +0.55)
3.14 갭상승 까마귀형, Upside Gap Two Crows (↑60%, ↓40%, +0.3)
3.15 하락장악확인형, Three Outside Down (↑31%, ↓69%, -0.38)
3.16 상승장악확인형, Three Outside Up (↑75%, ↓25%, +0.5)
3.17 상승잉태확인형, Three Inside Up (↑65%, ↓35%, +0.3)
3.18 폭발십자캔들팬턴, Collapsing Doji (↑37%, ↓63%, -0.26)
3.19 하락갭삼법, Downside Gap Three Methods (↑62%, ↓38%, +0.24)

end.

https://github.com/suparjotamin/stockie

https://towardsdatascience.com/how-to-identify-japanese-candlesticks-patterns-in-python-b835d1cc72f7

https://medium.com/analytics-vidhya/recognizing-over-50-candlestick-patterns-with-python-4f02a1822cb5

https://www.skyer9.pe.kr/wordpress/?p=1773 

1. 기본형태

https://tickkle.tistory.com/entry/%EC%BA%94%EB%93%A4%EA%B8%B0%EB%B3%B8%ED%98%95%ED%83%9C?category=1025295 

2. 상승반전패턴

https://tickkle.tistory.com/entry/%EC%83%81%EC%8A%B9%EB%B0%98%EC%A0%84%ED%8C%A8%ED%84%B4?category=1025295 

3. 상승지속패턴

https://tickkle.tistory.com/entry/%EC%83%81%EC%8A%B9%EC%A7%80%EC%86%8D%ED%8C%A8%ED%84%B4?category=1025295 

4. 하락지속패턴

https://tickkle.tistory.com/entry/%ED%95%98%EB%9D%BD%EC%A7%80%EC%86%8D%ED%8C%A8%ED%84%B4?category=1025295 

5. 하락반전패턴

https://tickkle.tistory.com/entry/%ED%95%98%EB%9D%BD%EB%B0%98%EC%A0%84%ED%8C%A8%ED%84%B4?category=1025295 

end.

https://www.cpubenchmark.net/laptop.html

상관계수(r)와 유의확율(p-value)

1. Correlation (상관관계)
  - 상관관계는 두 변수간의 관계의 강도와 방향을 설명하는 척도입니다. 변수가 얼마나 밀접하게 관련되어 있는지 연구하는 데 사용되는 방법을 Correlation Analysis(상관분석)라 합니다.
   그러나, 상관관계가 인과 관계를 의미하지는 않습니다. 한 변수의 변경이 다른 변수의 변경과 연관되어 있다는 사실이 실제로 다른 변수가 변경되는 것을 의미하지는 않습니다. (인관관계는 선형회귀분석으로)

2. 상관계수(r)와 상관 관계 해석
2.1 상관계수 r : 두 연속 변수 간의 연관 정도에 대한 수치.
  - 계수값은 항상 -1과 1사이이며, 변수간의 선형 관계의 강도와 방향을 모두 측정함.
 1) Strength(힘): 계수의 절대 값이 클수록 관계가 강해집니다.
   - 0에서 +1, -1 사이의 값은 약함, 보통 및 강함 관계의 척도를 나타냅니다. 마찬가지로 R은 -1 또는 1 에 가까울 수록 관계의 강도가 증가합니다.
 2) Direction(방향): 계수 기호(더하기, 빼기)는 관계의 방향을 나타냅니다.
    - 양의 계수는 직접적인 상관 관계를 나타내며 한 변수가 증가하면 다른 변수도 증가합니다.
    - 음의 계수는 역 상관을 나타내며, 그래프에서 하향 기울기를 생성합니다. 한 변수가 증가면 다른 변수는 감소하는 경향이 있습니다.
     . 계수 1은 완전환 양의 관계를 의미합니다. 한 변수가 증가하면 다른 변수는 비례 적으로 증가합니다.
     . 계수 -1은 완벽한 음의 관계를 의미합2니다. 한 변수가 증가하면 다른 변수는 비례적으로 감사합니다.
     . 계수 0은 두 변수간에 관계가 없음을 의미합니다. 데이터 포인트는 그래프 전체에 흩어져 있습니다.

3. 관련수식
3.1 R : 상관계수
3.2 R Square(R제곱값: R^2) : 결정계수
   - 이 값은 추세선이 데이터에 얼마나 잘 대응하는 지 나타내며, R^2가 1에 가까울수록 적합도가 더 좋습니다.
3.3 P-Value
  - 유의확률. 귀무가설이 맞다고 가정할 때 얻은 결과보다 극단적인 결과가 실제로 관측될 확률이다. p-value가 작을 수록 그 정도가 약하다고 보며, 특정 값(대게 0.05나 0.01)보다 작을 경우 귀무가설을 기각하는 것이나 여기에는 여러가지 문제들이 있다.
  - p-value는 보통 검정통계량(t)과 자유도(df) 로 계산됨. (엑셀에서 TDIS(t, df, false(양측검증)) 함수)
  - 검정통계량(t)은 상관계수에서 아래 식을 통해서 계산할 수 있음.
  - 자유도는 표본쌍(n) - 2로 계산
  ps1. 결정계수 = 상관계수^2
  ps2. 상관계수 = SQRT(결정계수) = 결정계수^0.5=√ℛ
  ps3. 차트에 추세선 추가하고, 상단에 추세선 수식 표시. (예: y= 0.003x + 11.11 (추세선 선형 방정식 y=ax + b))

3.4 선형방정식
  : 차트에서 추세선을 그리기위한 계산식. X의 min/max로 추세선을 추가하고, 산술식을 표기한다.
 1) 선형방정식
  Y = {0}X (+or-) {1}
    - {0} : slope
    - {1}: yintercept

end.