選對統計檢定
Python 的實驗設計
James Chapman
Curriculum Manager, DataCamp
如何選擇合適的檢定
- 資料集特徵
- 資料型別
- 分配 → 許多檢定假設為常態！
- 變數數量
- 假設
結果：結論更準確、可靠！
- t 檢定、ANOVA、卡方檢定
資料集：運動表現
- 訓練課程與飲食對運動表現的影響
athletic_perf.sample(n=5)
Athlete_ID Training_Program Diet_Type Initial_Fitness Performance_Inc
167 Endurance Plant-Based High 9.113040
289 Endurance Keto Low 11.039744
164 Endurance Plant-Based Medium 11.614835
30 Strength Keto Medium 7.384686
186 HIIT High-Protein Low 6.776078
獨立樣本 t 檢定
- 比較「兩組」的平均數
- 假設：常態分配、變異數相等
from scipy.stats import ttest_ind group1 = athletic_perf[athletic_perf['Training_Program'] == 'HIIT']['Performance_Inc'] group2 = athletic_perf[athletic_perf['Training_Program'] == 'Endurance']['Performance_Inc']t_stat, p_val = ttest_ind(group1, group2) print(f"T-statistic: {t_stat}, P-value: {p_val}")
T-statistic: 0.20671020082911742, P-value: 0.8364563849070663
p_val > $\alpha$ → 證據不足，無平均數差異
單因子 ANOVA
- 比較「多組（>2）」的平均數
- 假設：各組變異數相等
from scipy.stats import f_oneway program_types = ['HIIT', 'Endurance', 'Strength'] groups = [athletic_perf_data[athletic_perf_data['Training_Program'] == program] ['Performance_Increase'] for program in program_types]f_stat, p_val = f_oneway(*groups) print(f"F-statistic: {f_stat}, P-value: {p_val}")
F-statistic: 1.5270022393256704, P-value: 0.2188859009050602
p_val > $\alpha$ → 證據不足，無平均數差異
卡方關聯檢定
- 檢驗「類別變數」之間的關聯
- 不需分配假設
from scipy.stats import chi2_contingency
import pandas as pd
contingency_table = pd.crosstab(athletic_perf['Training_Program'],
athletic_perf['Diet_Type'])
Diet_Type High-Protein Keto Plant-Based
Training_Program
Endurance 33 28 33
HIIT 27 32 40
Strength 38 29 40
卡方關聯檢定
chi2_stat, p_val, dof, expected = chi2_contingency(contingency_table)
print(f"Chi2-statistic: {chi2_stat}, P-value: {p_val}")
Chi2-statistic: 2.154450885821988, P-value: 0.7073764021451127
p_val > $\alpha$ → 證據不足，無關聯
一起來練習吧！
Python 的實驗設計
