The Pearl (Page 7)

The Pearl

Sign in Subscribe

**TAIEX.s43_Training.3th.RandomForest+SVR+XGB+Transformer. 訓練4

**TAIEX.s43_Training.3th.RandomForest+SVR+XGB+Transformer. 訓練4

Picture 1 Shapes: test_labels: (6540,) transformer_predictions: (6540, 2) Adjusted transformer_predictions shape: (6540,) Random Forest MSE: 2538928.7593811294 SVR MSE: 13866232.38063333 XGBoost MSE: 5859048.980611635 LSTM MSE: 1862506.720476739 Transformer MSE: 0.020672246942703427 import os import pandas as pd import numpy as np import matplotlib.pyplot as

**TAIEX.s43_Training.3th.RandomForest+SVR+XGB+Transformer. 訓練3

**TAIEX.s43_Training.3th.RandomForest+SVR+XGB+Transformer. 訓練3

transformer_predictions: (6540, 2) Adjusted transformer_predictions shape: (6540,) Random Forest MSE: 2539254.211486759 SVR MSE: 13866232.398817357 XGBoost MSE: 5859048.979625798 LSTM MSE: 1753263.4853189574 Transformer MSE: 2732.7598143748187 import os import pandas as pd import numpy as np import matplotlib.pyplot as plt import matplotlib.dates as mdates

**TAIEX.s42_Training.2nd.RandomForest+SVR+XGB+Transformer. 簡易訓練

**TAIEX.s42_Training.2nd.RandomForest+SVR+XGB+Transformer. 簡易訓練

import os import pandas as pd import numpy as np import matplotlib.pyplot as plt import matplotlib.dates as mdates from sklearn.preprocessing import StandardScaler from sklearn.ensemble import RandomForestRegressor from sklearn.svm import SVR import tensorflow as tf from tensorflow.keras.models import Sequential from tensorflow.keras.layers import

TAIEX.s41_Training.1'st.第一次訓練,RandomForest+SVR

TAIEX.s41_Training.1'st.第一次訓練,RandomForest+SVR

import pandas as pd import numpy as np import matplotlib.pyplot as plt import matplotlib.dates as mdates from sklearn.preprocessing import StandardScaler from sklearn.ensemble import RandomForestRegressor from sklearn.svm import SVR # 加載數據 file_path = '/content/drive/My Drive/MSCI_Taiwan_30_data_with_OBV.csv' data

**TAIEX.s30_Future_Engineering.特徵工程

**TAIEX.s30_Future_Engineering.特徵工程

import pandas as pd import numpy as np from sklearn.preprocessing import StandardScaler # 加載數據 file_path = '/content/drive/My Drive/MSCI_Taiwan_30_data_with_OBV.csv' data = pd.read_csv(file_path) # 更改列名 data.rename(columns={'Close_MSCI': 'Close'}, inplace=True) # 1. 計算移動平均線

**TAIEX.ML.s20_Missing Values Analysis.缺失值分析(填補重製)

**TAIEX.ML.s20_Missing Values Analysis.缺失值分析(填補重製)

# Missing values analysis missing_values = data.isnull().sum() print("Missing Values Analysis:") print(missing_values) Missing Values Analysis: Date 0 ST_Code 0 ST_Name 0 Open 0 High 0 Low 0 Close_MSCI 0 Adj_Close 0 Volume 0 MA7 180 MA21 600 MA50 1470 MA100 2970

**TAIEX.ML.s14_Box Plot.箱型圖

**TAIEX.ML.s14_Box Plot.箱型圖

Box Plot Matrix Map 提供了對各變數數據分佈的直觀了解，有助於識別數據中的異常值和分佈特徵。通過觀察這些圖表，可以為後續的數據清洗和特徵工程提供參考。 Box Plot Matrix Map 圖表顯示了每個變數的數據分佈情況，其中包括中位數、四分位數範圍、異常值等信息。以下是圖表的摘要說明： 1. 數據範圍： * 大多數變數的數據範圍較小，集中在較低的數值區間。 * Volume 和 OBV 的數據範圍明顯大於其他變數，因此在圖表中顯示時有較大的差異。 2. 異常值（Outliers）： * 幾乎所有變數都存在異常值，這些異常值顯示為圖表中的小圓點。 * 特別是 Volume、Band Width、MACD Line、Signal Line 等變數，異常值較多且分佈較廣。 3. 變數分佈： * 部分變數（如 Aroon Up、Aroon Down、RSI7、

**TAIEX.ML.s13_Normality_Test.常態分配檢定

**TAIEX.ML.s13_Normality_Test.常態分配檢定

正常性檢驗結果顯示所有變數的 p 值均為 0，表明數據顯著偏離正態分佈。這是正常的結果，因為股票市場數據通常不遵循正態分佈。 * P 值 (p-value)：在統計檢驗中，P 值用於衡量觀察結果與零假設（通常是數據符合某種分佈，比如正態分佈）的偏離程度。當 P 值小於某個顯著性水平（如 0.05）時，我們通常拒絕零假設。 * 正常性檢驗：常見的正態性檢驗方法包括 Shapiro-Wilk 檢驗。在這種檢驗中，零假設是數據來自正態分佈。在我們的檢驗中，所有變數的 P 值均為 0（或極小），這意味著我們有足夠的證據拒絕數據來自正態分佈的假設。這在股票市場數據中是正常現象，因為這些數據往往具有尖峰厚尾（leptokurtic）或偏態（skewness），不符合正態分佈。 # Sample data sampled_data = data.sample(n=

**TAIEX.ML.s12_Correlation Heatmap.熱點圖(1)

**TAIEX.ML.s12_Correlation Heatmap.熱點圖(1)

# Select only numeric columns for correlation analysis numeric_data = data.select_dtypes(include=[float, int]) # Correlation analysis correlation_matrix = numeric_data.corr() print("Correlation Matrix:") print(correlation_matrix) # Plot correlation heatmap plt.figure(figsize=(16, 12)) sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm') plt.title(

**TAIEX.ML.s11_Scatter Plot.散布圖(2)

**TAIEX.ML.s11_Scatter Plot.散布圖(2)

# Sample data sampled_data = data.sample(n=1000, random_state=42) # Plot pairwise relationships sns.pairplot(sampled_data) plt.suptitle('Scatter Plot of Variables', y=1.02) plt.show()

**TAIEX.ML.s11_Scatter Plot.散布圖(1)

**TAIEX.ML.s11_Scatter Plot.散布圖(1)

'**有些變數的視覺化趨勢就蠻明顯的(見下圖) import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns from scipy.stats import pearsonr # Mount Google Drive from google.colab import drive drive.mount('/content/drive', force_remount=True) # Load the data file_path = '/content/drive/

TAIEX.ML.s02_Dataset.數據描述

TAIEX.ML.s02_Dataset.數據描述

from google.colab import drive drive.mount('/content/drive') import pandas as pd import matplotlib.pyplot as plt import seaborn as sns from scipy import stats # Load the data file_path = '/content/drive/My Drive/MSCI_Taiwan_30_data_with_OBV.csv' data = pd.read_csv(

TAIEX.ML.s01_Varians.變數說明

TAIEX.ML.s01_Varians.變數說明

變數名稱說明 Close_TAIEX TAIEX 指數的收盤價 Adj_Close 收盤價 Aroon Down Aroon 指標中的下降線 Aroon Up Aroon 指標中的上升線 Band Width 布林帶的寬度 Beta_120 過去 120 天的 Beta 值 Beta_60 過去 60 天的 Beta 值 CCI20 20 天商品通道指數 Close 收盤價 CMO14 14 天的錢德動量震盪指標 %D 隨機指標中的 %D 線 Date 日期 High 最高價

**Scatter Matrix by Varians and ST_Stock ( P<0.05)

**Scatter Matrix by Varians and ST_Stock ( P<0.05)

Mounted at /content/drive Overall Pearson Correlation with Close_TAIEX: Pearson Correlation P-value RSI21 0.185104 2.388407e-245 RSI14 0.151384 6.581929e-165 CMO14 0.098249 4.022700e-70 Aroon Up 0.095180 3.432801e-65 RSI7 0.094756 9.565736e-66 CCI20 0.087107 3.717489e-55 Signal Line 0.083004 1.343307e-49

Scatter Matrix by ST_Stock.

Scatter Matrix by ST_Stock.

Pearson Correlation with Close_TAIEX for 2330.TW (TSMC) Pearson Correlation P-value Close_TAIEX_TAIEX 1.000000 0.000000e+00 Adj_Close 0.954742 0.000000e+00 Low 0.951000 0.000000e+00 MA7 0.950148 0.000000e+00 Middle Band 0.949835 0.000000e+00 MA21 0.949652 0.

Close_TAIEX -N.Scatter Plot Matrix_0630

Close_TAIEX -N.Scatter Plot Matrix_0630

import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns from scipy.stats import pearsonr # Mount Google Drive from google.colab import drive drive.mount('/content/drive', force_remount=True) # Load the data file_path = '/content/drive/My Drive/MSCI_

Heat MAP.MSCI 個股對 TAIEX 的 Pearson analysis.(f)0630

Heat MAP.MSCI 個股對 TAIEX 的 Pearson analysis.(f)0630

MSCI 個股對 TAIEX 的 Pearson analysis import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns from scipy.stats import pearsonr # Mount Google Drive from google.colab import drive drive.mount('/content/drive', force_remount=True) # Load the data file_path

MSCI.TW30.Pearson correlation coefficient,Draft,發散的亂算

import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns from scipy.stats import pearsonr # Mount Google Drive from google.colab import drive drive.mount('/content/drive', force_remount=True) # Load the data file_path = '/content/drive/My Drive/MSCI_

盛夏月亮

Date: 2024-06-30 夏夜星稀月更明笑靨如花水清波影清雅如蘭悄然秀芳華馥郁時光且相迎情意滿懷溢心間芳華初綻放怎能不心傾盛夏時光如畫我的眼角是滿滿的驕傲且行且珍惜情意滿懷溢心間芳華初綻放怎能不心傾盛夏時光如畫景帶著滿滿的驕傲這裡總會有最溫暖的守候盛夏時光歲月靜好小月光輝長照心間如含苞的花朵綻放無與倫比的美麗我的小月光我的小月亮

MSCI.TW30.Down&Characteristic variables.0630

MSCI.TW30.Down&Characteristic variables.0630

SVM.dft1.TAIEX.POC.0628

SVM.dft1.TAIEX.POC.0628

from google.colab import drive drive.mount('/content/drive') !pip install yfinance scikit-learn matplotlib statsmodels ta import yfinance as yf import pandas as pd import numpy as np import matplotlib.pyplot as plt from sklearn.preprocessing import MinMaxScaler from sklearn.svm import SVR from sklearn.decomposition import PCA

Random_Forest.(dft5).POC.0628

Random_forest(dft.4)_TAIEX_0628

Random_forest(dft.4)_TAIEX_0628

加入 beta,beta_120 beta_120的離散大 from google.colab import drive drive.mount('/content/drive') !pip install yfinance scikit-learn matplotlib statsmodels import yfinance as yf import pandas as pd import numpy as np import matplotlib.pyplot as plt from sklearn.preprocessing import MinMaxScaler from sklearn.ensemble import RandomForestRegressor

Random_Forest(dft.3)_TAIEX_POC_0628

Random_Forest(dft.3)_TAIEX_POC_0628

* n_iter=10：這意味著隨機搜索將嘗試 10 種不同的參數組合。您可以根據時間和資源的限制來調整這個數字。 * cv=3：使用 3 折交叉驗證來評估每個參數組合，這可以在保持模型性能評估的同時減少計算時間。 from google.colab import drive drive.mount('/content/drive') !pip install yfinance scikit-learn matplotlib statsmodels import yfinance as yf import pandas as pd import numpy as np import matplotlib.pyplot as plt from sklearn.preprocessing

如何獲取 TAIEX 成分股代碼

如何獲取 TAIEX 成分股代碼

!pip install pandas requests gspread oauth2client import pandas as pd import requests import gspread from google.colab import auth from oauth2client.client import GoogleCredentials from google.colab import drive # 掛載 Google Drive drive.mount('/content/drive') # 獲取 TAIEX 成分股代碼 url = "https://isin.twse.com.tw/isin/C_