def fit_disaggregation_model(model):
"""Disaggregation approach"""
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(name='dis', ages=model.ages,
knots=knots,
smoothing=pl.inf,
interpolation_method='linear')
## Disaggregate input data
a = []
p = []
n = []
for i in model.input_data.index:
a_s, a_e = model.input_data.ix[i, 'age_start'], model.input_data.ix[i, 'age_end']
a += range(a_s, a_e)
p += [model.input_data.ix[i, 'value']] * (a_e - a_s)
n += [float(model.input_data.ix[i, 'effective_sample_size']) / (a_e - a_s)] * (a_e - a_s)
a = pl.array(a)
p = pl.array(p)
n = pl.array(n)
model.vars['pi'] = mc.Lambda('pi_dis', lambda mu=model.vars['mu_age'], a=a: mu[a])
delta = mc.Uniform('delta_mid', 0., 1000., value=10.)
model.vars += {'delta': delta}
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(name='dis',
pi=model.vars['pi'],
delta=model.vars['delta'],
p=p, # TODO: change this parameter name to "r" to match the book chapter
n=n)
fit_model(model)
def fit_age_standardizing_model(model):
"""Age-standardizing model"""
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(name='std', ages=model.ages,
knots=knots,
smoothing=pl.inf,
interpolation_method='linear')
model.vars += dismod3.age_group.age_standardize_approx(name='std', ages=model.ages,
age_weights=pl.ones_like(model.ages),
mu_age=model.vars['mu_age'],
age_start=model.input_data['age_start'], age_end=model.input_data['age_end'])
delta = mc.Uniform('delta_mid', 0., 1000., value=10.)
model.vars += {'delta': delta}
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(name='std',
pi=model.vars['mu_interval'],
delta=model.vars['delta'],
p=model.input_data['value'],
n=model.input_data['effective_sample_size'])
fit_model(model)
def fit_age_integrating_model(model):
"""Age-integrating model"""
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(
name="std", ages=model.ages, knots=knots, smoothing=pl.inf, interpolation_method="linear"
)
model.vars += dismod3.age_group.age_integrate_approx(
name="int",
ages=model.ages,
age_weights=model.input_data["age_weights"],
mu_age=model.vars["mu_age"],
age_start=model.input_data["age_start"],
age_end=model.input_data["age_end"],
)
model.vars += {"delta": mc.Uniform("delta_std", 0.0, 1000.0, value=10.0)}
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(
name="std",
pi=model.vars["mu_interval"],
delta=model.vars["delta"],
p=model.input_data["value"],
n=model.input_data["effective_sample_size"],
)
fit_model(model)
def fit_midpoint_covariate_model(model):
"""Midpoint-covariate model"""
# Create age-group model
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(name='midc', ages=model.ages,
knots=knots,
smoothing=pl.inf,
interpolation_method='linear')
## Midpoint model to represent age-group data
model.vars += dismod3.age_group.midpoint_covariate_approx(name='midc', ages=model.ages,
mu_age=model.vars['mu_age'],
age_start=model.input_data['age_start'], age_end=model.input_data['age_end'])
## Uniform prior on negative binomial rate model over-dispersion term
delta = mc.Uniform('delta_mid', 0., 1000., value=10.)
model.vars += {'delta': delta}
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(name='midc',
pi=model.vars['mu_interval'],
delta=model.vars['delta'],
p=model.input_data['value'],
n=model.input_data['effective_sample_size'])
fit_model(model)
def fit_disaggregation_model(model):
"""Disaggregation approach"""
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(
name="dis", ages=model.ages, knots=knots, smoothing=pl.inf, interpolation_method="linear"
)
## Disaggregate input data
a = []
p = []
n = []
for i in model.input_data.index:
a_s, a_e = model.input_data.ix[i, "age_start"], model.input_data.ix[i, "age_end"]
a += range(a_s, a_e)
p += [model.input_data.ix[i, "value"]] * (a_e - a_s)
n += [float(model.input_data.ix[i, "effective_sample_size"]) / (a_e - a_s)] * (a_e - a_s)
a = pl.array(a)
p = pl.array(p)
n = pl.array(n)
model.vars["pi"] = mc.Lambda("pi_dis", lambda mu=model.vars["mu_age"], a=a: mu[a])
model.vars["delta"] = mc.Uniform("delta_dis", 0.0, 1000.0, value=10.0)
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(
name="dis",
pi=model.vars["pi"],
delta=model.vars["delta"],
p=p, # TODO: change this parameter name to "r" to match the book chapter
n=n,
)
fit_model(model)
def fit_alt_midpoint_covariate_model(model):
"""Midpoint/covariate model sq"""
# Create age-group model
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(
name="midc", ages=model.ages, knots=knots, smoothing=pl.inf, interpolation_method="linear"
)
## Midpoint model to represent age-group data
model.vars += dismod3.age_group.midpoint_covariate_approx(
name="midc",
ages=model.ages,
mu_age=model.vars["mu_age"],
age_start=model.input_data["age_start"],
age_end=model.input_data["age_end"],
transform=lambda x: x ** 2.0,
)
## Uniform prior on negative binomial rate model over-dispersion term
model.vars += {"delta": mc.Uniform("delta_midc", 0.0, 1000.0, value=10.0)}
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(
name="midc",
pi=model.vars["mu_interval"],
delta=model.vars["delta"],
p=model.input_data["value"],
n=model.input_data["effective_sample_size"],
)
fit_model(model)
def fit_age_integrating_model(model):
"""Age-integrating model"""
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(name='std',
ages=model.ages,
knots=knots,
smoothing=pl.inf,
interpolation_method='linear')
model.vars += dismod3.age_group.age_integrate_approx(
name='int',
ages=model.ages,
age_weights=model.input_data['age_weights'],
mu_age=model.vars['mu_age'],
age_start=model.input_data['age_start'],
age_end=model.input_data['age_end'])
model.vars += {'delta': mc.Uniform('delta_std', 0., 1000., value=10.)}
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(
name='std',
pi=model.vars['mu_interval'],
delta=model.vars['delta'],
p=model.input_data['value'],
n=model.input_data['effective_sample_size'])
fit_model(model)
def fit_alt_midpoint_covariate_model(model):
"""Midpoint/covariate model sq"""
# Create age-group model
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(name='midc',
ages=model.ages,
knots=knots,
smoothing=pl.inf,
interpolation_method='linear')
## Midpoint model to represent age-group data
model.vars += dismod3.age_group.midpoint_covariate_approx(
name='midc',
ages=model.ages,
mu_age=model.vars['mu_age'],
age_start=model.input_data['age_start'],
age_end=model.input_data['age_end'],
transform=lambda x: x**2.)
## Uniform prior on negative binomial rate model over-dispersion term
model.vars += {'delta': mc.Uniform('delta_midc', 0., 1000., value=10.)}
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(
name='midc',
pi=model.vars['mu_interval'],
delta=model.vars['delta'],
p=model.input_data['value'],
n=model.input_data['effective_sample_size'])
fit_model(model)
def fit_disaggregation_model(model):
"""Disaggregation approach"""
## Spline model to represent age-specific rate
model.vars += dismod3.age_pattern.spline(name='dis',
ages=model.ages,
knots=knots,
smoothing=pl.inf,
interpolation_method='linear')
## Disaggregate input data
a = []
p = []
n = []
for i in model.input_data.index:
a_s, a_e = model.input_data.ix[i, 'age_start'], model.input_data.ix[
i, 'age_end']
a += range(a_s, a_e)
p += [model.input_data.ix[i, 'value']] * (a_e - a_s)
n += [
float(model.input_data.ix[i, 'effective_sample_size']) /
(a_e - a_s)
] * (a_e - a_s)
a = pl.array(a)
p = pl.array(p)
n = pl.array(n)
model.vars['pi'] = mc.Lambda('pi_dis',
lambda mu=model.vars['mu_age'], a=a: mu[a])
delta = mc.Uniform('delta_mid', 0., 1000., value=10.)
model.vars += {'delta': delta}
## Negative binomial rate model
model.vars += dismod3.rate_model.neg_binom(
name='dis',
pi=model.vars['pi'],
delta=model.vars['delta'],
p=p, # TODO: change this parameter name to "r" to match the book chapter
n=n)
fit_model(model)
