为什么Python中的t-test(scipy,statsmodels)会给出与R,Stata或Excel不同的结果？

(问题已解决; x,y和s1,s2的大小不同)

在R：

x <- c(373,398,245,272,238,241,134,410,158,125,198,252,577,208,260)
y <- c(411,471,320,364,311,390,163,424,228,144,246,371,680,384,279,303)
t.test(x,y)
t = -1.6229,df = 29.727,p-value = 0.1152

在STATA和Excel中获得相同的数字

t.test(x,y,alternative="less")
t = -1.6229,p-value = 0.05758

无论我尝试哪种选项,我都无法使用statsmodels.stats.weightstats.ttest_ind或scipy.stats.ttest_ind复制相同的结果.

statsmodels.stats.weightstats.ttest_ind(s1,s2,alternative="two-sided",usevar="unequal")
(-1.8912081781378358,0.066740317997990656,35.666557473974343)

scipy.stats.ttest_ind(s1,equal_var=False)
(array(-1.8912081781378338),0.066740317997990892)

scipy.stats.ttest_ind(s1,equal_var=True)
(array(-1.8912081781378338),0.066664507499812745)

必须有成千上万的人使用Python来计算t检验.我们都得到不正确的结果吗？ (我通常依赖Python,但这次我用STATA检查了我的结果).

最佳答案

这是我得到的结果,默认等于var：

>>> x_ = (373,260)
>>> y_ = (411,303)

>>> from scipy import stats
>>> stats.ttest_ind(x_,y_)
(array(-1.62292672368488),0.11506840827144681)

>>> import statsmodels.api as sm
>>> sm.stats.ttest_ind(x_,y_)
(-1.6229267236848799,0.11506840827144681,30.0)

和不等var：

>>> statsmodels.stats.weightstats.ttest_ind(x_,y_,usevar="unequal")
(-1.6229267236848799,0.11516398707890187,29.727196553288369)
>>> stats.ttest_ind(x_,equal_var=False)
(array(-1.62292672368488),0.11516398707890187)