Python中常用的统计检验代码分享

import scipy.stats as stats

# 生成两组样本数据
data1 = [25, 30, 35, 40, 45]
data2 = [30, 32, 35, 38, 42]

# 执行独立样本 t 检验
t_stat, p_value = stats.ttest_ind(data1, data2)
print("t 统计量:", t_stat)
print("p 值:", p_value)

if p_value < 0.05:
    print("差异显著")
else:
    print("差异不显著")

import scipy.stats as stats

# 创建观察频数表
observed = [[30, 15], [25, 20]]

# 执行卡方检验
chi2, p_value, _, _ = stats.chi2_contingency(observed)
print("卡方统计量:", chi2)
print("p 值:", p_value)

if p_value < 0.05:
    print("差异显著")
else:
    print("差异不显著")

import scipy.stats as stats

# 创建两组连续变量数据
x = [1, 2, 3, 4, 5]
y = [2, 4, 5, 4, 6]

# 计算 Pearson 相关系数
correlation, _ = stats.pearsonr(x, y)
print("Pearson 相关系数:", correlation)

if abs(correlation) > 0.7:
    print("线性相关性强")
else:
    print("线性相关性弱")

import scipy.stats as stats

# 单样本数据
data = [28, 30, 29, 32, 31]

# 已知均值
known_mean = 30

# 执行单样本 t 检验
t_stat, p_value = stats.ttest_1samp(data, known_mean)
print("t 统计量:", t

_stat)
print("p 值:", p_value)

if p_value < 0.05:
    print("差异显著")
else:
    print("差异不显著")

import scipy.stats as stats

# 创建两组配对样本数据
before = [28, 30, 29, 32, 31]
after = [27, 29, 28, 31, 30]

# 执行 Wilcoxon 符号秩检验
w_stat, p_value = stats.wilcoxon(before, after)
print("Wilcoxon 符号秩统计量:", w_stat)
print("p 值:", p_value)

if p_value < 0.05:
    print("差异显著")
else:
    print("差异不显著")

import scipy.stats as stats

# 创建多组样本数据
group1 = [65, 68, 72, 70, 74]
group2 = [58, 63, 65, 61, 59]
group3 = [72, 70, 75, 71, 73]

# 执行单因素 ANOVA
f_stat, p_value = stats.f_oneway(group1, group2, group3)
print("F 统计量:", f_stat)
print("p 值:", p_value)

if p_value < 0.05:
    print("差异显著")
else:
    print("差异不显著")

import scipy.stats as stats

# 创建两组数据
data1 = [0.1, 0.2, 0.3, 0.4, 0.5]
data2 = [0.2, 0.3, 0.4, 0.5, 0.6]

# 执行 Kolmogorov-Smirnov 检验
ks_stat, p_value = stats.ks_2samp(data1, data2)
print("KS 统计量:", ks_stat)
print("p 值:", p_value)

if p_value < 0.05:
    print("差异显著")
else:
    print("差异不显著")

import scipy.stats as stats

# 创建样本数据
data = [2.3, 3.```python
7, 4.5, 6.8, 5.1, 8.2]

# 执行 Shapiro-Wilk 正态性检验
statistic, p_value = stats.shapiro(data)
print("Shapiro-Wilk 统计量:", statistic)
print("p 值:", p_value)

if p_value < 0.05:
    print("不符合正态分布")
else:
    print("符合正态分布")

import scipy.stats as stats

# 创建自变量和因变量数据
x = [1, 2, 3, 4, 5]
y = [2, 4, 5, 4, 6]

# 执行线性回归分析
slope, intercept, r_value, p_value, std_err = stats.linregress(x, y)
print("斜率:", slope)
print("截距:", intercept)
print("相关系数:", r_value)
print("p 值:", p_value)

if p_value < 0.05:
    print("回归模型显著")
else:
    print("回归模型不显著")

import statsmodels.api as sm

# 创建自变量和因变量数据
x = [1, 2, 3, 4, 5]
y = [2, 4, 5, 4, 6]

# 添加常数项
x = sm.add_constant(x)

# 执行分位数回归分析
quantiles = [0.25, 0.5, 0.75]
for q in quantiles:
    model = sm.QuantReg(y, x)
    result = model.fit(q=q)
    print(f"分位数 {q}:")
    print(result.summary())