M-Estimators for Robust Linear Modeling¶

In [ ]:
%matplotlib inline

from __future__ import print_function
from statsmodels.compat import lmap
import numpy as np
from scipy import stats
import matplotlib.pyplot as plt

import statsmodels.api as sm

• An M-estimator minimizes the function

$$Q(e_i, \rho) = \sum_i~\rho \left (\frac{e_i}{s}\right )$$

where $\rho$ is a symmetric function of the residuals

• The effect of $\rho$ is to reduce the influence of outliers
• $s$ is an estimate of scale.
• The robust estimates $\hat{\beta}$ are computed by the iteratively re-weighted least squares algorithm
• We have several choices available for the weighting functions to be used
In [ ]:
norms = sm.robust.norms

In [ ]:
def plot_weights(support, weights_func, xlabels, xticks):
fig = plt.figure(figsize=(12,8))
ax.plot(support, weights_func(support))
ax.set_xticks(xticks)
ax.set_xticklabels(xlabels, fontsize=16)
ax.set_ylim(-.1, 1.1)
return ax


Andrew's Wave¶

In [ ]:
help(norms.AndrewWave.weights)

In [ ]:
a = 1.339
support = np.linspace(-np.pi*a, np.pi*a, 100)
andrew = norms.AndrewWave(a=a)
plot_weights(support, andrew.weights, ['$-\pi*a$', '0', '$\pi*a$'], [-np.pi*a, 0, np.pi*a]);


Hampel's 17A¶

In [ ]:
help(norms.Hampel.weights)

In [ ]:
c = 8
support = np.linspace(-3*c, 3*c, 1000)
hampel = norms.Hampel(a=2., b=4., c=c)
plot_weights(support, hampel.weights, ['3*c', '0', '3*c'], [-3*c, 0, 3*c]);


Huber's t¶

In [ ]:
help(norms.HuberT.weights)

In [ ]:
t = 1.345
support = np.linspace(-3*t, 3*t, 1000)
huber = norms.HuberT(t=t)
plot_weights(support, huber.weights, ['-3*t', '0', '3*t'], [-3*t, 0, 3*t]);


Least Squares¶

In [ ]:
help(norms.LeastSquares.weights)

In [ ]:
support = np.linspace(-3, 3, 1000)
lst_sq = norms.LeastSquares()
plot_weights(support, lst_sq.weights, ['-3', '0', '3'], [-3, 0, 3]);


Ramsay's Ea¶

In [ ]:
help(norms.RamsayE.weights)

In [ ]:
a = .3
support = np.linspace(-3*a, 3*a, 1000)
ramsay = norms.RamsayE(a=a)
plot_weights(support, ramsay.weights, ['-3*a', '0', '3*a'], [-3*a, 0, 3*a]);


Trimmed Mean¶

In [ ]:
help(norms.TrimmedMean.weights)

In [ ]:
c = 2
support = np.linspace(-3*c, 3*c, 1000)
trimmed = norms.TrimmedMean(c=c)
plot_weights(support, trimmed.weights, ['-3*c', '0', '3*c'], [-3*c, 0, 3*c]);


Tukey's Biweight¶

In [ ]:
help(norms.TukeyBiweight.weights)

In [ ]:
c = 4.685
support = np.linspace(-3*c, 3*c, 1000)
tukey = norms.TukeyBiweight(c=c)
plot_weights(support, tukey.weights, ['-3*c', '0', '3*c'], [-3*c, 0, 3*c]);


Scale Estimators¶

• Robust estimates of the location
In [ ]:
x = np.array([1, 2, 3, 4, 500])

• The mean is not a robust estimator of location
In [ ]:
x.mean()

• The median, on the other hand, is a robust estimator with a breakdown point of 50%
In [ ]:
np.median(x)

• Analagously for the scale
• The standard deviation is not robust
In [ ]:
x.std()


Median Absolute Deviation

$$median_i |X_i - median_j(X_j)|)$$

Standardized Median Absolute Deviation is a consistent estimator for $\hat{\sigma}$

$$\hat{\sigma}=K \cdot MAD$$

where $K$ depends on the distribution. For the normal distribution for example,

$$K = \Phi^{-1}(.75)$$

In [ ]:
stats.norm.ppf(.75)

In [ ]:
print(x)

In [ ]:
sm.robust.scale.mad(x)

In [ ]:
np.array([1,2,3,4,5.]).std()

• The default for Robust Linear Models is MAD
• another popular choice is Huber's proposal 2
In [ ]:
np.random.seed(12345)
fat_tails = stats.t(6).rvs(40)

In [ ]:
kde = sm.nonparametric.KDEUnivariate(fat_tails)
kde.fit()
fig = plt.figure(figsize=(12,8))
ax.plot(kde.support, kde.density);

In [ ]:
print(fat_tails.mean(), fat_tails.std())

In [ ]:
print(stats.norm.fit(fat_tails))

In [ ]:
print(stats.t.fit(fat_tails, f0=6))

In [ ]:
huber = sm.robust.scale.Huber()
loc, scale = huber(fat_tails)
print(loc, scale)

In [ ]:
sm.robust.mad(fat_tails)

In [ ]:
sm.robust.mad(fat_tails, c=stats.t(6).ppf(.75))

In [ ]:
sm.robust.scale.mad(fat_tails)


Duncan's Occupational Prestige data - M-estimation for outliers¶

In [ ]:
from statsmodels.graphics.api import abline_plot
from statsmodels.formula.api import ols, rlm

In [ ]:
prestige = sm.datasets.get_rdataset("Duncan", "car", cache=True).data

In [ ]:
print(prestige.head(10))

In [ ]:
fig = plt.figure(figsize=(12,12))
ax1.scatter(prestige.income, prestige.prestige)
xy_outlier = prestige.loc['minister', ['income','prestige']]
ax1.annotate('Minister', xy_outlier, xy_outlier+1, fontsize=16)
ylabel='Prestige')
ax2.scatter(prestige.education, prestige.prestige);

In [ ]:
ols_model = ols('prestige ~ income + education', prestige).fit()
print(ols_model.summary())

In [ ]:
infl = ols_model.get_influence()
student = infl.summary_frame()['student_resid']
print(student)

In [ ]:
print(student.loc[np.abs(student) > 2])

In [ ]:
print(infl.summary_frame().loc['minister'])

In [ ]:
sidak = ols_model.outlier_test('sidak')
print(sidak)

In [ ]:
fdr = ols_model.outlier_test('fdr_bh')
print(fdr)

In [ ]:
rlm_model = rlm('prestige ~ income + education', prestige).fit()
print(rlm_model.summary())

In [ ]:
print(rlm_model.weights)


Hertzprung Russell data for Star Cluster CYG 0B1 - Leverage Points¶

• Data is on the luminosity and temperature of 47 stars in the direction of Cygnus.
In [ ]:
dta = sm.datasets.get_rdataset("starsCYG", "robustbase", cache=True).data

In [ ]:
from matplotlib.patches import Ellipse
fig = plt.figure(figsize=(12,8))
ax = fig.add_subplot(111, xlabel='log(Temp)', ylabel='log(Light)', title='Hertzsprung-Russell Diagram of Star Cluster CYG OB1')
ax.scatter(*dta.values.T)
# highlight outliers
e = Ellipse((3.5, 6), .2, 1, alpha=.25, color='r')
ax.annotate('Red giants', xy=(3.6, 6), xytext=(3.8, 6),
arrowprops=dict(facecolor='black', shrink=0.05, width=2),
horizontalalignment='left', verticalalignment='bottom',
clip_on=True, # clip to the axes bounding box
fontsize=16,
)
# annotate these with their index
for i,row in dta.loc[dta['log.Te'] < 3.8].iterrows():
ax.annotate(i, row, row + .01, fontsize=14)
xlim, ylim = ax.get_xlim(), ax.get_ylim()

In [ ]:
from IPython.display import Image
Image(filename='star_diagram.png')

In [ ]:
y = dta['log.light']
ols_model = sm.OLS(y, X).fit()
abline_plot(model_results=ols_model, ax=ax)

In [ ]:
rlm_mod = sm.RLM(y, X, sm.robust.norms.TrimmedMean(.5)).fit()
abline_plot(model_results=rlm_mod, ax=ax, color='red')

• Why? Because M-estimators are not robust to leverage points.
In [ ]:
infl = ols_model.get_influence()

In [ ]:
h_bar = 2*(ols_model.df_model + 1 )/ols_model.nobs
hat_diag = infl.summary_frame()['hat_diag']
hat_diag.loc[hat_diag > h_bar]

In [ ]:
sidak2 = ols_model.outlier_test('sidak')
print(sidak2)

In [ ]:
fdr2 = ols_model.outlier_test('fdr_bh')
print(fdr2)

• Let's delete that line
In [ ]:
l =  ax.lines[-1]
l.remove()
del l

In [ ]:
weights = np.ones(len(X))
weights[X[X['log.Te'] < 3.8].index.values - 1] = 0
wls_model = sm.WLS(y, X, weights=weights).fit()
abline_plot(model_results=wls_model, ax=ax, color='green')

• MM estimators are good for this type of problem, unfortunately, we don't yet have these yet.
• It's being worked on, but it gives a good excuse to look at the R cell magics in the notebook.
In [ ]:
yy = y.values[:,None]
xx = X['log.Te'].values[:,None]

In [ ]:
%load_ext rpy2.ipython

In [ ]:
%R library(robustbase)
%Rpush yy xx
%R mod <- lmrob(yy ~ xx);
%R params <- mod\$coefficients;
%Rpull params

In [ ]:
%R print(mod)

In [ ]:
print(params)

In [ ]:
abline_plot(intercept=params[0], slope=params[1], ax=ax, color='red')


Exercise: Breakdown points of M-estimator¶

In [ ]:
np.random.seed(12345)
nobs = 200
beta_true = np.array([3, 1, 2.5, 3, -4])
X = np.random.uniform(-20,20, size=(nobs, len(beta_true)-1))
# stack a constant in front
X = sm.add_constant(X, prepend=True) # np.c_[np.ones(nobs), X]
mc_iter = 500
contaminate = .25 # percentage of response variables to contaminate

In [ ]:
all_betas = []
for i in range(mc_iter):
y = np.dot(X, beta_true) + np.random.normal(size=200)
random_idx = np.random.randint(0, nobs, size=int(contaminate * nobs))
y[random_idx] = np.random.uniform(-750, 750)
beta_hat = sm.RLM(y, X).fit().params
all_betas.append(beta_hat)

In [ ]:
all_betas = np.asarray(all_betas)
se_loss = lambda x : np.linalg.norm(x, ord=2)**2
se_beta = lmap(se_loss, all_betas - beta_true)


Squared error loss¶

In [ ]:
np.array(se_beta).mean()

In [ ]:
all_betas.mean(0)

In [ ]:
beta_true

In [ ]:
se_loss(all_betas.mean(0) - beta_true)