Thống kê Mô tả - Độ đo Tập trung

1import numpy as np
2
3# Dữ liệu
4scores = [7, 8, 6, 9, 7, 8, 10, 5]
5
6# Mean đơn giản
7mean_score = np.mean(scores)
8print(f"Mean: {mean_score}")  # 7.5
9
10# Weighted Mean
11grades = [8, 7, 9]
12credits = [4, 3, 2]
13weighted_mean = np.average(grades, weights=credits)
14print(f"Weighted Mean: {weighted_mean:.2f}")  # 7.89

1import numpy as np
2
3# n lẻ
4data_odd = [3, 7, 2, 9, 4]
5print(f"Median (n lẻ): {np.median(data_odd)}")  # 4.0
6
7# n chẵn
8data_even = [3, 7, 2, 9, 4, 6]
9print(f"Median (n chẵn): {np.median(data_even)}")  # 5.0

1from scipy import stats
2import numpy as np
3
4# Unimodal
5data = [1, 2, 2, 3, 4]
6mode_result = stats.mode(data, keepdims=True)
7print(f"Mode: {mode_result.mode[0]}")  # 2
8print(f"Count: {mode_result.count[0]}")  # 2
9
10# Sử dụng pandas cho multiple modes
11import pandas as pd
12data_bimodal = pd.Series([1, 1, 2, 2, 3])
13print(f"Modes: {data_bimodal.mode().tolist()}")  # [1, 2]

1import numpy as np
2import matplotlib.pyplot as plt
3from scipy import stats
4
5# Tạo dữ liệu với outlier
6data_normal = [10, 12, 11, 13, 14, 12, 11, 13]
7data_with_outlier = [10, 12, 11, 13, 14, 12, 11, 100]
8
9# So sánh
10print("=== Data without outlier ===")
11print(f"Mean: {np.mean(data_normal):.2f}")
12print(f"Median: {np.median(data_normal):.2f}")
13
14print("\n=== Data with outlier ===")
15print(f"Mean: {np.mean(data_with_outlier):.2f}")
16print(f"Median: {np.median(data_with_outlier):.2f}")
17
18# Visualization
19fig, axes = plt.subplots(1, 2, figsize=(12, 4))
20
21# Normal data
22axes[0].hist(data_normal, bins=10, edgecolor='black', alpha=0.7)
23axes[0].axvline(np.mean(data_normal), color='red', linestyle='--', label=f'Mean: {np.mean(data_normal):.1f}')
24axes[0].axvline(np.median(data_normal), color='green', linestyle='--', label=f'Median: {np.median(data_normal):.1f}')
25axes[0].set_title('Without Outlier')
26axes[0].legend()
27
28# Data with outlier
29axes[1].hist(data_with_outlier, bins=10, edgecolor='black', alpha=0.7)
30axes[1].axvline(np.mean(data_with_outlier), color='red', linestyle='--', label=f'Mean: {np.mean(data_with_outlier):.1f}')
31axes[1].axvline(np.median(data_with_outlier), color='green', linestyle='--', label=f'Median: {np.median(data_with_outlier):.1f}')
32axes[1].set_title('With Outlier')
33axes[1].legend()
34
35plt.tight_layout()
36plt.show()