Linear Mixed Effects Models

In [1]:
%matplotlib inline

import numpy as np
import statsmodels.api as sm
import statsmodels.formula.api as smf
/build/statsmodels-GE7Zhw/statsmodels-0.8.0/.pybuild/cpython3_3.6_statsmodels/build/statsmodels/compat/pandas.py:56: FutureWarning: The pandas.core.datetools module is deprecated and will be removed in a future version. Please use the pandas.tseries module instead.
  from pandas.core import datetools
In [2]:
%load_ext rpy2.ipython
---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
<ipython-input-2-f4f84eb1b14e> in <module>()
----> 1 get_ipython().magic('load_ext rpy2.ipython')

/usr/lib/python3/dist-packages/IPython/core/interactiveshell.py in magic(self, arg_s)
   2158         magic_name, _, magic_arg_s = arg_s.partition(' ')
   2159         magic_name = magic_name.lstrip(prefilter.ESC_MAGIC)
-> 2160         return self.run_line_magic(magic_name, magic_arg_s)
   2161 
   2162     #-------------------------------------------------------------------------

/usr/lib/python3/dist-packages/IPython/core/interactiveshell.py in run_line_magic(self, magic_name, line)
   2079                 kwargs['local_ns'] = sys._getframe(stack_depth).f_locals
   2080             with self.builtin_trap:
-> 2081                 result = fn(*args,**kwargs)
   2082             return result
   2083 

<decorator-gen-63> in load_ext(self, module_str)

/usr/lib/python3/dist-packages/IPython/core/magic.py in <lambda>(f, *a, **k)
    186     # but it's overkill for just that one bit of state.
    187     def magic_deco(arg):
--> 188         call = lambda f, *a, **k: f(*a, **k)
    189 
    190         if callable(arg):

/usr/lib/python3/dist-packages/IPython/core/magics/extension.py in load_ext(self, module_str)
     35         if not module_str:
     36             raise UsageError('Missing module name.')
---> 37         res = self.shell.extension_manager.load_extension(module_str)
     38 
     39         if res == 'already loaded':

/usr/lib/python3/dist-packages/IPython/core/extensions.py in load_extension(self, module_str)
     81             if module_str not in sys.modules:
     82                 with prepended_to_syspath(self.ipython_extension_dir):
---> 83                     __import__(module_str)
     84             mod = sys.modules[module_str]
     85             if self._call_load_ipython_extension(mod):

ModuleNotFoundError: No module named 'rpy2'
In [3]:
%R library(lme4)
UsageError: Line magic function `%R` not found.

Comparing R lmer to Statsmodels MixedLM

The Statsmodels imputation of linear mixed models (MixedLM) closely follows the approach outlined in Lindstrom and Bates (JASA 1988). This is also the approach followed in the R package LME4. Other packages such as Stata, SAS, etc. should also be consistent with this approach, as the basic techniques in this area are mostly mature.

Here we show how linear mixed models can be fit using the MixedLM procedure in Statsmodels. Results from R (LME4) are included for comparison.

Here are our import statements:

Growth curves of pigs

These are longitudinal data from a factorial experiment. The outcome variable is the weight of each pig, and the only predictor variable we will use here is "time". First we fit a model that expresses the mean weight as a linear function of time, with a random intercept for each pig. The model is specified using formulas. Since the random effects structure is not specified, the default random effects structure (a random intercept for each group) is automatically used.

In [4]:
data = sm.datasets.get_rdataset('dietox', 'geepack').data
md = smf.mixedlm("Weight ~ Time", data, groups=data["Pig"])
mdf = md.fit()
print(mdf.summary())
---------------------------------------------------------------------------
gaierror                                  Traceback (most recent call last)
/usr/lib/python3.6/urllib/request.py in do_open(self, http_class, req, **http_conn_args)
   1317                 h.request(req.get_method(), req.selector, req.data, headers,
-> 1318                           encode_chunked=req.has_header('Transfer-encoding'))
   1319             except OSError as err: # timeout error

/usr/lib/python3.6/http/client.py in request(self, method, url, body, headers, encode_chunked)
   1238         """Send a complete request to the server."""
-> 1239         self._send_request(method, url, body, headers, encode_chunked)
   1240 

/usr/lib/python3.6/http/client.py in _send_request(self, method, url, body, headers, encode_chunked)
   1284             body = _encode(body, 'body')
-> 1285         self.endheaders(body, encode_chunked=encode_chunked)
   1286 

/usr/lib/python3.6/http/client.py in endheaders(self, message_body, encode_chunked)
   1233             raise CannotSendHeader()
-> 1234         self._send_output(message_body, encode_chunked=encode_chunked)
   1235 

/usr/lib/python3.6/http/client.py in _send_output(self, message_body, encode_chunked)
   1025         del self._buffer[:]
-> 1026         self.send(msg)
   1027 

/usr/lib/python3.6/http/client.py in send(self, data)
    963             if self.auto_open:
--> 964                 self.connect()
    965             else:

/usr/lib/python3.6/http/client.py in connect(self)
   1391 
-> 1392             super().connect()
   1393 

/usr/lib/python3.6/http/client.py in connect(self)
    935         self.sock = self._create_connection(
--> 936             (self.host,self.port), self.timeout, self.source_address)
    937         self.sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)

/usr/lib/python3.6/socket.py in create_connection(address, timeout, source_address)
    703     err = None
--> 704     for res in getaddrinfo(host, port, 0, SOCK_STREAM):
    705         af, socktype, proto, canonname, sa = res

/usr/lib/python3.6/socket.py in getaddrinfo(host, port, family, type, proto, flags)
    744     addrlist = []
--> 745     for res in _socket.getaddrinfo(host, port, family, type, proto, flags):
    746         af, socktype, proto, canonname, sa = res

gaierror: [Errno -2] Name or service not known

During handling of the above exception, another exception occurred:

URLError                                  Traceback (most recent call last)
<ipython-input-4-48520994a507> in <module>()
----> 1 data = sm.datasets.get_rdataset('dietox', 'geepack').data
      2 md = smf.mixedlm("Weight ~ Time", data, groups=data["Pig"])
      3 mdf = md.fit()
      4 print(mdf.summary())

/build/statsmodels-GE7Zhw/statsmodels-0.8.0/.pybuild/cpython3_3.6_statsmodels/build/statsmodels/datasets/utils.py in get_rdataset(dataname, package, cache)
    288                      "master/doc/"+package+"/rst/")
    289     cache = _get_cache(cache)
--> 290     data, from_cache = _get_data(data_base_url, dataname, cache)
    291     data = read_csv(data, index_col=0)
    292     data = _maybe_reset_index(data)

/build/statsmodels-GE7Zhw/statsmodels-0.8.0/.pybuild/cpython3_3.6_statsmodels/build/statsmodels/datasets/utils.py in _get_data(base_url, dataname, cache, extension)
    219     url = base_url + (dataname + ".%s") % extension
    220     try:
--> 221         data, from_cache = _urlopen_cached(url, cache)
    222     except HTTPError as err:
    223         if '404' in str(err):

/build/statsmodels-GE7Zhw/statsmodels-0.8.0/.pybuild/cpython3_3.6_statsmodels/build/statsmodels/datasets/utils.py in _urlopen_cached(url, cache)
    210     # not using the cache or didn't find it in cache
    211     if not from_cache:
--> 212         data = urlopen(url).read()
    213         if cache is not None:  # then put it in the cache
    214             _cache_it(data, cache_path)

/usr/lib/python3.6/urllib/request.py in urlopen(url, data, timeout, cafile, capath, cadefault, context)
    221     else:
    222         opener = _opener
--> 223     return opener.open(url, data, timeout)
    224 
    225 def install_opener(opener):

/usr/lib/python3.6/urllib/request.py in open(self, fullurl, data, timeout)
    524             req = meth(req)
    525 
--> 526         response = self._open(req, data)
    527 
    528         # post-process response

/usr/lib/python3.6/urllib/request.py in _open(self, req, data)
    542         protocol = req.type
    543         result = self._call_chain(self.handle_open, protocol, protocol +
--> 544                                   '_open', req)
    545         if result:
    546             return result

/usr/lib/python3.6/urllib/request.py in _call_chain(self, chain, kind, meth_name, *args)
    502         for handler in handlers:
    503             func = getattr(handler, meth_name)
--> 504             result = func(*args)
    505             if result is not None:
    506                 return result

/usr/lib/python3.6/urllib/request.py in https_open(self, req)
   1359         def https_open(self, req):
   1360             return self.do_open(http.client.HTTPSConnection, req,
-> 1361                 context=self._context, check_hostname=self._check_hostname)
   1362 
   1363         https_request = AbstractHTTPHandler.do_request_

/usr/lib/python3.6/urllib/request.py in do_open(self, http_class, req, **http_conn_args)
   1318                           encode_chunked=req.has_header('Transfer-encoding'))
   1319             except OSError as err: # timeout error
-> 1320                 raise URLError(err)
   1321             r = h.getresponse()
   1322         except:

URLError: <urlopen error [Errno -2] Name or service not known>

Here is the same model fit in R using LMER:

In [5]:
%%R 
data(dietox, package='geepack')
UsageError: Cell magic `%%R` not found.
In [6]:
%R print(summary(lmer('Weight ~ Time + (1|Pig)', data=dietox)))
UsageError: Line magic function `%R` not found.

Note that in the Statsmodels summary of results, the fixed effects and random effects parameter estimates are shown in a single table. The random effect for animal is labeled "Intercept RE" in the Statmodels output above. In the LME4 output, this effect is the pig intercept under the random effects section.

There has been a lot of debate about whether the standard errors for random effect variance and covariance parameters are useful. In LME4, these standard errors are not displayed, because the authors of the package believe they are not very informative. While there is good reason to question their utility, we elected to include the standard errors in the summary table, but do not show the corresponding Wald confidence intervals.

Next we fit a model with two random effects for each animal: a random intercept, and a random slope (with respect to time). This means that each pig may have a different baseline weight, as well as growing at a different rate. The formula specifies that "Time" is a covariate with a random coefficient. By default, formulas always include an intercept (which could be suppressed here using "0 + Time" as the formula).

In [7]:
md = smf.mixedlm("Weight ~ Time", data, groups=data["Pig"], re_formula="~Time")
mdf = md.fit()
print(mdf.summary())
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-7-1ecbbd21c6e0> in <module>()
----> 1 md = smf.mixedlm("Weight ~ Time", data, groups=data["Pig"], re_formula="~Time")
      2 mdf = md.fit()
      3 print(mdf.summary())

NameError: name 'data' is not defined

Here is the same model fit using LMER in R:

In [8]:
%R print(summary(lmer("Weight ~ Time + (1 + Time | Pig)", data=dietox)))
UsageError: Line magic function `%R` not found.

The random intercept and random slope are only weakly correlated $(0.294 / \sqrt{19.493 * 0.416} \approx 0.1)$. So next we fit a model in which the two random effects are constrained to be uncorrelated:

In [9]:
.294 / (19.493 * .416)**.5
Out[9]:
0.10324316832591753
In [10]:
md = smf.mixedlm("Weight ~ Time", data, groups=data["Pig"],
                  re_formula="~Time")
free = sm.regression.mixed_linear_model.MixedLMParams.from_components(np.ones(2), 
                                                                      np.eye(2))

mdf = md.fit(free=free)
print(mdf.summary())
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-10-ee7e60cd7cf0> in <module>()
----> 1 md = smf.mixedlm("Weight ~ Time", data, groups=data["Pig"],
      2                   re_formula="~Time")
      3 free = sm.regression.mixed_linear_model.MixedLMParams.from_components(np.ones(2), 
      4                                                                       np.eye(2))
      5 

NameError: name 'data' is not defined

The likelihood drops by 0.3 when we fix the correlation parameter to 0. Comparing 2 x 0.3 = 0.6 to the chi^2 1 df reference distribution suggests that the data are very consistent with a model in which this parameter is equal to 0.

Here is the same model fit using LMER in R (note that here R is reporting the REML criterion instead of the likelihood, where the REML criterion is twice the log likeihood):

In [11]:
%R print(summary(lmer("Weight ~ Time + (1 | Pig) + (0 + Time | Pig)", data=dietox)))
UsageError: Line magic function `%R` not found.

Sitka growth data

This is one of the example data sets provided in the LMER R library. The outcome variable is the size of the tree, and the covariate used here is a time value. The data are grouped by tree.

In [12]:
data = sm.datasets.get_rdataset("Sitka", "MASS").data
endog = data["size"]
data["Intercept"] = 1
exog = data[["Intercept", "Time"]]
---------------------------------------------------------------------------
gaierror                                  Traceback (most recent call last)
/usr/lib/python3.6/urllib/request.py in do_open(self, http_class, req, **http_conn_args)
   1317                 h.request(req.get_method(), req.selector, req.data, headers,
-> 1318                           encode_chunked=req.has_header('Transfer-encoding'))
   1319             except OSError as err: # timeout error

/usr/lib/python3.6/http/client.py in request(self, method, url, body, headers, encode_chunked)
   1238         """Send a complete request to the server."""
-> 1239         self._send_request(method, url, body, headers, encode_chunked)
   1240 

/usr/lib/python3.6/http/client.py in _send_request(self, method, url, body, headers, encode_chunked)
   1284             body = _encode(body, 'body')
-> 1285         self.endheaders(body, encode_chunked=encode_chunked)
   1286 

/usr/lib/python3.6/http/client.py in endheaders(self, message_body, encode_chunked)
   1233             raise CannotSendHeader()
-> 1234         self._send_output(message_body, encode_chunked=encode_chunked)
   1235 

/usr/lib/python3.6/http/client.py in _send_output(self, message_body, encode_chunked)
   1025         del self._buffer[:]
-> 1026         self.send(msg)
   1027 

/usr/lib/python3.6/http/client.py in send(self, data)
    963             if self.auto_open:
--> 964                 self.connect()
    965             else:

/usr/lib/python3.6/http/client.py in connect(self)
   1391 
-> 1392             super().connect()
   1393 

/usr/lib/python3.6/http/client.py in connect(self)
    935         self.sock = self._create_connection(
--> 936             (self.host,self.port), self.timeout, self.source_address)
    937         self.sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)

/usr/lib/python3.6/socket.py in create_connection(address, timeout, source_address)
    703     err = None
--> 704     for res in getaddrinfo(host, port, 0, SOCK_STREAM):
    705         af, socktype, proto, canonname, sa = res

/usr/lib/python3.6/socket.py in getaddrinfo(host, port, family, type, proto, flags)
    744     addrlist = []
--> 745     for res in _socket.getaddrinfo(host, port, family, type, proto, flags):
    746         af, socktype, proto, canonname, sa = res

gaierror: [Errno -2] Name or service not known

During handling of the above exception, another exception occurred:

URLError                                  Traceback (most recent call last)
<ipython-input-12-2f7a7b6b4426> in <module>()
----> 1 data = sm.datasets.get_rdataset("Sitka", "MASS").data
      2 endog = data["size"]
      3 data["Intercept"] = 1
      4 exog = data[["Intercept", "Time"]]

/build/statsmodels-GE7Zhw/statsmodels-0.8.0/.pybuild/cpython3_3.6_statsmodels/build/statsmodels/datasets/utils.py in get_rdataset(dataname, package, cache)
    288                      "master/doc/"+package+"/rst/")
    289     cache = _get_cache(cache)
--> 290     data, from_cache = _get_data(data_base_url, dataname, cache)
    291     data = read_csv(data, index_col=0)
    292     data = _maybe_reset_index(data)

/build/statsmodels-GE7Zhw/statsmodels-0.8.0/.pybuild/cpython3_3.6_statsmodels/build/statsmodels/datasets/utils.py in _get_data(base_url, dataname, cache, extension)
    219     url = base_url + (dataname + ".%s") % extension
    220     try:
--> 221         data, from_cache = _urlopen_cached(url, cache)
    222     except HTTPError as err:
    223         if '404' in str(err):

/build/statsmodels-GE7Zhw/statsmodels-0.8.0/.pybuild/cpython3_3.6_statsmodels/build/statsmodels/datasets/utils.py in _urlopen_cached(url, cache)
    210     # not using the cache or didn't find it in cache
    211     if not from_cache:
--> 212         data = urlopen(url).read()
    213         if cache is not None:  # then put it in the cache
    214             _cache_it(data, cache_path)

/usr/lib/python3.6/urllib/request.py in urlopen(url, data, timeout, cafile, capath, cadefault, context)
    221     else:
    222         opener = _opener
--> 223     return opener.open(url, data, timeout)
    224 
    225 def install_opener(opener):

/usr/lib/python3.6/urllib/request.py in open(self, fullurl, data, timeout)
    524             req = meth(req)
    525 
--> 526         response = self._open(req, data)
    527 
    528         # post-process response

/usr/lib/python3.6/urllib/request.py in _open(self, req, data)
    542         protocol = req.type
    543         result = self._call_chain(self.handle_open, protocol, protocol +
--> 544                                   '_open', req)
    545         if result:
    546             return result

/usr/lib/python3.6/urllib/request.py in _call_chain(self, chain, kind, meth_name, *args)
    502         for handler in handlers:
    503             func = getattr(handler, meth_name)
--> 504             result = func(*args)
    505             if result is not None:
    506                 return result

/usr/lib/python3.6/urllib/request.py in https_open(self, req)
   1359         def https_open(self, req):
   1360             return self.do_open(http.client.HTTPSConnection, req,
-> 1361                 context=self._context, check_hostname=self._check_hostname)
   1362 
   1363         https_request = AbstractHTTPHandler.do_request_

/usr/lib/python3.6/urllib/request.py in do_open(self, http_class, req, **http_conn_args)
   1318                           encode_chunked=req.has_header('Transfer-encoding'))
   1319             except OSError as err: # timeout error
-> 1320                 raise URLError(err)
   1321             r = h.getresponse()
   1322         except:

URLError: <urlopen error [Errno -2] Name or service not known>

Here is the statsmodels LME fit for a basic model with a random intercept. We are passing the endog and exog data directly to the LME init function as arrays. Also note that endog_re is specified explicitly in argument 4 as a random intercept (although this would also be the default if it were not specified).

In [13]:
md = sm.MixedLM(endog, exog, groups=data["tree"], exog_re=exog["Intercept"])
mdf = md.fit()
print(mdf.summary())
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-13-e6b2b322447f> in <module>()
----> 1 md = sm.MixedLM(endog, exog, groups=data["tree"], exog_re=exog["Intercept"])
      2 mdf = md.fit()
      3 print(mdf.summary())

NameError: name 'endog' is not defined

Here is the same model fit in R using LMER:

In [14]:
%%R
data(Sitka, package="MASS")
print(summary(lmer("size ~ Time + (1 | tree)", data=Sitka)))
UsageError: Cell magic `%%R` not found.

We can now try to add a random slope. We start with R this time. From the code and output below we see that the REML estimate of the variance of the random slope is nearly zero.

In [15]:
%R print(summary(lmer("size ~ Time + (1 + Time | tree)", data=Sitka)))
UsageError: Line magic function `%R` not found.

If we run this in statsmodels LME with defaults, we see that the variance estimate is indeed very small, which leads to a warning about the solution being on the boundary of the parameter space. The regression slopes agree very well with R, but the likelihood value is much higher than that returned by R.

In [16]:
exog_re = exog.copy()
md = sm.MixedLM(endog, exog, data["tree"], exog_re)
mdf = md.fit()
print(mdf.summary())
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-16-cdffaa9ce318> in <module>()
----> 1 exog_re = exog.copy()
      2 md = sm.MixedLM(endog, exog, data["tree"], exog_re)
      3 mdf = md.fit()
      4 print(mdf.summary())

NameError: name 'exog' is not defined

We can further explore the random effects struture by constructing plots of the profile likelihoods. We start with the random intercept, generating a plot of the profile likelihood from 0.1 units below to 0.1 units above the MLE. Since each optimization inside the profile likelihood generates a warning (due to the random slope variance being close to zero), we turn off the warnings here.

In [17]:
import warnings

with warnings.catch_warnings():
    warnings.filterwarnings("ignore")
    likev = mdf.profile_re(0, 're', dist_low=0.1, dist_high=0.1)
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-17-cc5407d2eb25> in <module>()
      3 with warnings.catch_warnings():
      4     warnings.filterwarnings("ignore")
----> 5     likev = mdf.profile_re(0, 're', dist_low=0.1, dist_high=0.1)

NameError: name 'mdf' is not defined

Here is a plot of the profile likelihood function. We multiply the log-likelihood difference by 2 to obtain the usual $\chi^2$ reference distribution with 1 degree of freedom.

In [18]:
import matplotlib.pyplot as plt
In [19]:
plt.figure(figsize=(10,8))
plt.plot(likev[:,0], 2*likev[:,1])
plt.xlabel("Variance of random slope", size=17)
plt.ylabel("-2 times profile log likelihood", size=17)
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-19-01cce210d83b> in <module>()
      1 plt.figure(figsize=(10,8))
----> 2 plt.plot(likev[:,0], 2*likev[:,1])
      3 plt.xlabel("Variance of random slope", size=17)
      4 plt.ylabel("-2 times profile log likelihood", size=17)

NameError: name 'likev' is not defined
<Figure size 720x576 with 0 Axes>

Here is a plot of the profile likelihood function. The profile likelihood plot shows that the MLE of the random slope variance parameter is a very small positive number, and that there is low uncertainty in this estimate.

In [20]:
re = mdf.cov_re.iloc[1, 1]
likev = mdf.profile_re(1, 're', dist_low=.5*re, dist_high=0.8*re)

plt.figure(figsize=(10, 8))
plt.plot(likev[:,0], 2*likev[:,1])
plt.xlabel("Variance of random slope", size=17)
plt.ylabel("-2 times profile log likelihood", size=17)
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-20-dc9b55e857c0> in <module>()
----> 1 re = mdf.cov_re.iloc[1, 1]
      2 likev = mdf.profile_re(1, 're', dist_low=.5*re, dist_high=0.8*re)
      3 
      4 plt.figure(figsize=(10, 8))
      5 plt.plot(likev[:,0], 2*likev[:,1])

NameError: name 'mdf' is not defined