繁体   English   中英

如何使用 statsmodels 获得未缩放的回归系数误差?

[英]How to get the unscaled regression coefficients errors using statsmodels?

我正在尝试使用 statsmodels 计算回归的系数误差。 也称为参数估计值的标准误差。 但我需要计算他们的“未缩放”版本。 我只设法用 NumPy 做到了。

您可以在文档中看到“未缩放”的含义: https://numpy.org/doc/stable/reference/generated/numpy.polyfit.html

cov bool or str, optional

    If given and not False, return not just the estimate but also its covariance matrix.
    By default, the covariance are scaled by chi2/dof, where dof = M - (deg + 1),
    i.e., the weights are presumed to be unreliable except in a relative sense and
    everything is scaled such that the reduced chi2 is unity. This scaling is omitted
    if cov='unscaled', as is relevant for the case that the weights are w = 1/sigma, with
    sigma known to be a reliable estimate of the uncertainty.

我正在使用此数据运行本文中代码的 rest:

import numpy as np
x = np.array([-0.841, -0.399, 0.599, 0.203, 0.527, 0.129, 0.703, 0.503])
y = np.array([1.01, 1.24, 1.09, 0.95, 1.02, 0.97, 1.01, 0.98])
sigmas = np.array([6872.26, 80.71, 47.97, 699.94, 57.55, 1561.54, 311.98, 501.08])
# The convention for weights are different 
sm_weights = np.array([1.0/sigma**2 for sigma in sigmas])
np_weights = np.array([1.0/sigma for sigma in sigmas])

使用 NumPy:

coefficients, cov = np.polyfit(x, y, deg=2, w=np_weights, cov='unscaled')
# The errors I need to get
print(np.sqrt(np.diag(cov))) # [917.57938013 191.2100413  211.29028248]

如果我使用 statsmodels 计算回归:

from sklearn.preprocessing import PolynomialFeatures
import statsmodels.api as smapi

polynomial_features = PolynomialFeatures(degree=2)
polynomial = polynomial_features.fit_transform(x.reshape(-1, 1))
model = smapi.WLS(y, polynomial, weights=sm_weights)
regression = model.fit()

# Get coefficient errors
# Notice the [::-1], statsmodels returns the coefficients in the reverse order NumPy does
print(regression.bse[::-1]) # [0.24532856, 0.05112286, 0.05649161]

所以我得到的值是不同的,但相关:

np_errors = np.sqrt(np.diag(cov))
sm_errors = regression.bse[::-1]
print(np_errors / sm_errors) # [3740.2061481, 3740.2061481, 3740.2061481]

NumPy 文档说the covariance are scaled by chi2/dof where dof = M - (deg + 1) 所以我尝试了以下方法:

degree = 2
model_predictions = np.polyval(coefficients, x)
residuals = (model_predictions - y)
chi_squared = np.sum(residuals**2)
degrees_of_freedom = len(x) - (degree + 1)
scale_factor = chi_squared / degrees_of_freedom

sm_cov = regression.cov_params()
unscaled_errors = np.sqrt(np.diag(sm_cov * scale_factor))[::-1] # [0.09848423, 0.02052266, 0.02267789]
unscaled_errors = np.sqrt(np.diag(sm_cov / scale_factor))[::-1] # [0.61112427, 0.12734931, 0.14072311]

我注意到我从 NumPy 得到的协方差矩阵比我从 statsmodels 得到的协方差矩阵大得多:

>>> cov
array([[ 841951.9188366 , -154385.61049538, -188456.18957375],
       [-154385.61049538,   36561.27989418,   31208.76422516],
       [-188456.18957375,   31208.76422516,   44643.58346933]])
>>> regression.cov_params()
array([[ 0.0031913 ,  0.00223093, -0.0134716 ],
       [ 0.00223093,  0.00261355, -0.0110361 ],
       [-0.0134716 , -0.0110361 ,  0.0601861 ]])

只要我不能使它们相等,我就不会得到相同的错误。 知道尺度上的差异意味着什么以及如何使两个协方差矩阵相等吗?

statsmodels 文档的某些部分组织得不好。 这是一个带有以下示例的笔记本https://www.statsmodels.org/devel/examples/notebooks/generated/chi2_fitting.html

OLS 和 WLS 等统计模型中的回归模型可以选择保持scale固定。 这相当于 numpy 和 scipy 中的cov="unscaled" 。statsmodels 选项更通用,因为它允许将比例固定为任何用户定义的值。

https://www.statsmodels.org/devel/generated/statsmodels.regression.linear_model.OLSResults.get_robustcov_results.html

我们有一个如示例中定义的 model,OLS 或 WLS,然后使用

regression = model.fit(cov_type="fixed scale")

将保持比例为 1,并且生成的协方差矩阵未缩放。

使用

regression = model.fit(cov_type="fixed scale", cov_kwds={"scale": 2})

将使比例固定在值二。

(相关讨论动机的一些链接在https://github.com/statsmodels/statsmodels/pull/2137

警告

固定尺度 cov_type 将用于基于参数估计值cov_params协方差的推论统计。 这会影响标准误差、t 检验、wald 检验以及置信区间和预测区间。

但是,某些其他结果统计可能不会调整为使用固定比例而不是估计比例,例如resid_pearson

https://github.com/statsmodels/statsmodels/issues/8190

暂无
暂无

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM