Pythonプログラミングで体感！
データサイエンス・エキスパート・ワークショップ研修
（※統計検定準1級過去問演習オンライン研修付帯可）

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import statsmodels.api as sm
from statsmodels.formula.api import ols

np.random.seed(None) # Each run will generate different random data
x_pos = np.random.rand(100) * 10
y_pos = 2.5 * x_pos + np.random.randn(100) * 5

x_neg = np.random.rand(100) * 10
y_neg = -2.5 * x_neg + np.random.randn(100) * 5

df_pos = pd.DataFrame({'x': x_pos, 'y': y_pos})
df_neg = pd.DataFrame({'x': x_neg, 'y': y_neg})

model_pos = ols('y ~ x', data=df_pos).fit()
intercept_pos, slope_pos = model_pos.params
r_squared_pos = model_pos.rsquared

model_neg = ols('y ~ x', data=df_neg).fit()
intercept_neg, slope_neg = model_neg.params
r_squared_neg = model_neg.rsquared

plt.figure(figsize=(20, 6))

plt.subplot(1, 2, 1)
plt.scatter(df_pos['x'], df_pos['y'], label='Data points')
plt.plot(df_pos['x'], intercept_pos + slope_pos * df_pos['x'], color='red', label=f'Regression line (R2={r_squared_pos:.2f})')
equation_pos = f'y = {slope_pos:.2f}x + {intercept_pos:.2f}'
plt.text(0.95, 0.05, equation_pos, transform=plt.gca().transAxes, fontsize=12, verticalalignment='bottom', horizontalalignment='right', bbox=dict(facecolor='white', alpha=0.5))
plt.xlabel('x')
plt.ylabel('y')
plt.title('Positive Correlation')
plt.legend(loc='upper left')
plt.grid(True)

plt.subplot(1, 2, 2)
plt.scatter(df_neg['x'], df_neg['y'], label='Data points')
plt.plot(df_neg['x'], intercept_neg + slope_neg * df_neg['x'], color='red', label=f'Regression line (R2={r_squared_neg:.2f})')
equation_neg = f'y = {slope_neg:.2f}x + {intercept_neg:.2f}'
plt.text(0.95, 0.05, equation_neg, transform=plt.gca().transAxes, fontsize=12, verticalalignment='bottom', horizontalalignment='right', bbox=dict(facecolor='white', alpha=0.5))
plt.xlabel('x')
plt.ylabel('y')
plt.title('Negative Correlation')
plt.legend(loc='upper left')
plt.grid(True)

plt.show()

anova_table_pos = sm.stats.anova_lm(model_pos, typ=2)
anova_table_neg = sm.stats.anova_lm(model_neg, typ=2)

print("ANOVA Table for Positive Correlation Data:")
print(anova_table_pos)
print("\nANOVA Table for Negative Correlation Data:")
print(anova_table_neg)

import numpy as np
import matplotlib.pyplot as plt
from scipy.stats import f

x = np.linspace(0, 5, 1000)

df1_values = [5, 6, 7, 8, 9]
df2_values = [2, 3, 4, 5, 6]

plt.figure(figsize=(10, 6))
for df1, df2 in zip(df1_values, df2_values):
　　y = f.pdf(x, df1, df2)
　　plt.plot(x, y, lw=2, label=f'df1={df1}, df2={df2}')

plt.title('Fixed F-Distributions')
plt.xlabel('x')
plt.ylabel('Density')
plt.legend()
plt.grid(True)
plt.show()