Python3.4.3的统计学拓展包
thippo
[email protected]
http://thippo.github.io
####必须依赖包
- numpy
- scipy
- pandas
- matplotlib
- scikit-learn
####安装
git clone [email protected]:thippo/Statistics_Python_Lib.git
cd Statistics_Python_Lib
python setup.py install
####模块结构
statspylib
ParameterEstimation
IntervalEstimate
DetermineSampleSize
HypothesisTesting
ParametricHypothesisTesting
NonparametricHypothesisTesting
VarianceAnalysis
ANOVA
LinearRegression
LinearRegression
MultipleLinearRegression
MultivariateStatistics
PCA
##- ParameterEstimation
###IntervalEstimate
单总体均值的区间估计(函数)
single_population_mean_interval_estimate(sample_list,confidence_level,population_standard_deviation)
单总体比例的区间估计(函数)
single_population_proportion_interval_estimate(sample_size,sample_proportion,confidence_level)
单总体方差的区间估计(函数)
single_population_variance_interval_estimate(sample_list,confidence_level)
双总体均值差的区间估计(函数)
double_population_mean_subtraction_interval_estimate(sampleA_list,sampleB_list,confidence_level,population_variance,populationA_variance='#',populationB_variance='#')
双总体均值差的区间估计:匹配样本(函数)
double_population_mean_subtraction_interval_estimate_matched_sample(sampleA_list,sampleB_list,confidence_level)
双总体比例差的区间估计(函数)
double_population_proportion_subtraction_interval_estimate(sampleA_size,sampleB_size,sampleA_proportion,sampleB_proportion,confidence_level)
双总体方差比的区间估计(函数)
double_population_variance_proportion_interval_estimate(sampleA_list,sampleB_list)
###DetermineSampleSize
估计总体均值时样本量的确定(函数)
determine_population_mean_sample_size(population_standard_deviation,confidence_level,estimate_error)
估计总体比例时样本量的确定(函数)
determine_population_proportion_sample_size(population_proportion,confidence_level,estimate_error)
##- HypothesisTesting
###ParametricHypothesisTesting
单总体均值的假设检验(函数)
single_population_mean_hypothesis_testing(test_direction,significant_level,sample_size,sample_mean,population_mean,population_standard_deviation,standard_deviation)
单总体比例的假设检验(函数)
single_population_proportion_hypothesis_testing(test_direction,significant_level,sample_size,sample_proportion,population_proportion)
单总体方差的假设检验(函数)
single_population_variance_hypothesis_testing(test_direction,significant_level,sample_size,sample_variance,population_variance)
双总体均值差的假设检验(函数)
double_population_mean_subtraction_hypothesis_testing(test_direction,significant_level,sampleA_size,sampleB_size,sampleA_mean,sampleB_mean,population_subtraction,population_variance,varianceA='#',varianceB='#')
双总体比例差的假设检验(函数)
double_population_proportion_subtraction_hypothesis_testing(test_direction,significant_level,sampleA_size,sampleB_size,sampleA_proportion,sampleB_proportion,population_proportion_subtraction)
双总体方差比的假设检验(函数)
double_population_variance_proportion_hypothesis_testing(test_direction,significant_level,sampleA_size,sampleB_size,sampleA_variance,sampleB_variance,population_variance_proportion)
双总体均值差的假设检验:匹配样本(函数)
double_population_mean_subtraction_hypothesis_testing_matched_sample(test_direction,significant_level,sample_size,sample_subtraction_mean,sample_subtraction_standard_deviation,population_subtraction_mean)
###NonparametricHypothesisTesting
单总体假设检验(类)
Single_Population_Test()
中位数符号检验(方法)
median_sign_test(arr,median)
Wilcoxon符号秩检验(方法)
wilcoxon_signed_rank_test(arr)
分布的一致性检验:卡方检验/适合度检验(函数)
goodness_of_fit_test(f_obs,f_exp,significance_level=0.05)
双总体假设检验(类)
Double_Population_Test()
卡方独立性检验(方法)不允许20%以下的格子的期望频数小于5
test_of_independence(observed,significance_level=0.05)
Fisher精确检验(方法)允许20%以下的格子的期望频数小于5
fisher_exact_test(table,alternative='two-sided')
位置参数的Wilcoxon秩和检验(方法)
wilcoxon_rank_sum_test(x,y)
位置参数的Mann-Whitney U检验(方法)
mann_whitney_U_test(x,y,use_continuity=True)
尺度参数的Mood检验(方法)
mood_test(x,y,axis=0)
尺度参数的Ansaru-Bradley检验(方法)
ansari_bradle_test(x,y)
多总体假设检验(类)
Multiple_Population_Test()
位置参数的Kruskal-Wallis秩和检验(方法)
kruskal_wallis_rank_sum_test(*arg)
尺度参数的Fligner-Killeen检验(方法)
fligner_killeen_test(*arg)
##- VarianceAnalysis
###ANOVA
单因素方差分析(类)
One_Way_ANOVA(dataframe)
方差齐性检验(方法)
homogeneity_of_variancet()
方差分析(方法)
ANOVA(significance_level=0.05,LSD=True)
双因素方差分析(类)
Two_Way_ANOVA(dataframe,interaction=False)
方差齐性检验(方法)
homogeneity_of_variancet()
方差分析(方法)
ANOVA(significance_level=0.05)
##- LinearRegression
###LinearRegression
一元线性回归(类)
One_Dimensional_Linear_Regression(x,y,significance_level=0.05)
线性回归分析(方法)
linear_regression(details=True,graphic=True)
显著性检验(方法)
significance_test()
点估计(方法)
point_estimate(point)
区间估计(方法)
interval_estimate(point)
残差分析(方法)
residual_analysis()
###MultipleLinearRegression
多元线性回归(类)
Multiple_Linear_Regression(x,y,significance_level=0.05)
线性回归分析(方法)
linear_regression()
共线性分析(方法)
multicollinearity()
预测(方法)
predict(x)
##- MultivariateStatistics
###PCA
主成分分析(函数)
PCA(dataMat,topNfeat, normalize=True)