def estimate_wrap(V, av, CHOICE, database, panelEst, ModNameSeqence, Draws):
'''
#
#
#
'''
if panelEst:
#database.panel('ID')
# Conditional to B_TIME_RND, the likelihood of one observation is
# given by the logit model (called the kernel)
obsprob = exp(models.loglogit(V, av, CHOICE))
# Conditional to B_TIME_RND, the likelihood of all observations for
# one individual (the trajectory) is the product of the likelihood of
# each observation.
condprobIndiv = PanelLikelihoodTrajectory(obsprob)
if Draws > 0:
condprobIndiv = MonteCarlo(condprobIndiv)
logprob = log(condprobIndiv)
else:
prob = exp(models.loglogit(V, av, CHOICE))
if Draws > 0:
prob = MonteCarlo(prob)
logprob = log(prob)
if Draws > 0:
biogeme = bio.BIOGEME(database, logprob, numberOfDraws=Draws)
else:
biogeme = bio.BIOGEME(database, logprob)
biogeme.modelName = ModNameSeqence
return biogeme.estimate()
def testEstimationAndSimulation(self):
biogeme = bio.BIOGEME(database, logprob, numberOfThreads=1)
results = biogeme.estimate()
self.assertAlmostEqual(results.data.logLike, -5214.049, 2)
biosim = bio.BIOGEME(database, simulate)
simresults = biosim.simulate(results.data.betaValues)
self.assertAlmostEqual(sum(simresults['Prob. train']),
888.3883902853023, 1)
self.assertAlmostEqual(sum(simresults['Elas. 1']), -17897.702976576973,
0)
def testEstimationAndSimulation(self):
biogeme = bio.BIOGEME(database, logprob)
results = biogeme.estimate()
self.assertAlmostEqual(results.data.logLike, -5214.049202307744, 1)
biosim = bio.BIOGEME(database, simulate)
simresults = biosim.simulate(results.data.betaValues)
self.assertAlmostEqual(sum(simresults['Prob. train']),
888.3883902853023, 1)
self.assertAlmostEqual(sum(simresults['Elas. 1']), -17898.420882021313,
0)
def evaluate_model(self, pandas_df_for_specified_country, model):
# Estimation of probabilities for each alternative on aggregate. Simulate / forecast.
def print_mode_shares(modename, modenumber):
seriesObj = simresults.apply(lambda x: True if x['Actual choice'] == modenumber else False, axis=1)
REAL = len(seriesObj[seriesObj == True].index)
seriesObj = simresults.apply(lambda x: True if x['Simulated choice'] == modenumber else False, axis=1)
SIMU = len(seriesObj[seriesObj == True].index)
shares = (modename, '--> Real:' + "{0:.1%}".format(REAL / simresults.shape[0]) +
'| Simu:' + "{0:.1%}".format(SIMU / simresults.shape[0]))
print(shares)
biosim = bio.BIOGEME(db.Database('estimationdb', pandas_df_for_specified_country), model.structure)
biosim.modelName = "simulated_model"
simresults = biosim.simulate(model.betas)
# Add a column containing the suggestion from the model
simresults['Simulated choice'] = simresults.idxmax(axis=1)
# Add a column containing the actual choice of the individual
simresults['Actual choice'] = pandas_df_for_specified_country['user_choice'].to_numpy()
# Add a column which compares the predicted against the RP choice (correct prediction = 1, wrong prediction = 0)
simresults['Correct prediction'] = np.where(simresults['Simulated choice'] == simresults['Actual choice'], 1, 0)
#print_mode_shares('Depart earlier', 1)
#print_mode_shares('Depart on-time', 2)
#print_mode_shares('Depart later ', 3)
return {'Model prediction accuracy': "{0:.1%}".format(simresults['Correct prediction'].mean()),
'Rho-square': "{0:.3}".format(model.results.getGeneralStatistics()['Rho-square-bar for the init. model'][0])}
def testSimulation(self):
biogeme = bio.BIOGEME(database,simulate)
biogeme.modelName = "01logit_simul"
results = biogeme.simulate()
self.assertAlmostEqual(sum(results['P1']),907.9992101964821,2)
self.assertAlmostEqual(sum(results['logit elas. 1']),-12673.838605478186,2)
self.assertAlmostEqual(sum(results['generic elas. 1']),-12673.838605478186,2)
def predict(self, trip_data, model_for_specified_country):
for i in range(1, 7):
trip_data['OCC_' + str(i)] = np.where(trip_data['user_occupation'] == i, 1, 0)
trip_data['AGE'] = self.__birthday_to_age(trip_data['user_birthday'])
# The trip is stored in a biogeme database, since it is required by Biogeme in order for it to function
tripdb = db.Database("SuggestionDB", trip_data)
# Simulate / forecast
biosuggest = bio.BIOGEME(tripdb, model_for_specified_country.structure)
biosuggest.modelName = "suggestion_to_user"
suggestionresults = biosuggest.simulate(model_for_specified_country.betas)
# Get the column index number of the max probability. This is My-TRAC's recommendation. Store it in a new col.
suggestionresults['Recommendation'] = suggestionresults.idxmax(axis=1)
suggestion = suggestionresults.values[0]
# print('Trip data = ', trip_data.to_json())
# print('Results = ',
# {'CAR': "{0:.1%}".format(suggestion[0]),
# 'PT': "{0:.1%}".format(suggestion[1]),
# 'BIKE/MOTO': "{0:.1%}".format(suggestion[2]),
# 'My-TRAC recommendation': int(suggestion[3])})
return {'mod_car': suggestion[0],
'mod_pt': suggestion[1],
'mod_motbike': suggestion[2]}
def testEstimation(self):
logprob = models.loglogit(V,av,CHOICE)
weight = 8.890991e-01 * (1.0 * (GROUP == 2) + 1.2 * (GROUP == 3))
formulas = {'loglike':logprob,'weight':weight}
biogeme = bio.BIOGEME(database,formulas)
results = biogeme.estimate()
self.assertAlmostEqual(results.data.logLike,-5273.743,2)
def predict(self, trip_data, model_for_specified_country):
trip_data['AGE'] = self.__birthday_to_age(trip_data['user_birthday'])
# The trip is stored in a biogeme database, since it is required by Biogeme in order for it to function
tripdb = db.Database("SuggestionDB", trip_data)
# Simulate / forecast
biosuggest = bio.BIOGEME(tripdb, model_for_specified_country.structure)
biosuggest.modelName = "suggestion_to_user"
suggestionresults = biosuggest.simulate(
model_for_specified_country.betas)
# Get the column index number of the max probability. This is My-TRAC's recommendation. Store it in a new col.
suggestionresults['Recommendation'] = suggestionresults.idxmax(axis=1)
suggestion = suggestionresults.values[0]
# print('Trip data = ', trip_data.to_json())
# print('Results = ',
# {'Depart earlier': "{0:.1%}".format(suggestion[0]),
# 'Depart on-time': "{0:.1%}".format(suggestion[1]),
# 'Depart later': "{0:.1%}".format(suggestion[2]),
# 'My-TRAC recommendation': int(suggestion[3])})
return {
'tod_earlier': suggestion[0],
'tod_ontime': suggestion[1],
'tod_later': suggestion[2]
}
def testEstimationScipy(self):
logprob = models.loglogit(V, av, CHOICE)
biogeme = bio.BIOGEME(database, logprob, seed=10, numberOfThreads=1)
biogeme.modelName = "test_01"
results = biogeme.estimate(bootstrap=10, algorithm=opt.scipy)
self.assertAlmostEqual(results.data.logLike, -5331.252, 2)
self.assertAlmostEqual(results.data.betas[0].bootstrap_stdErr,
0.04773096176427237, 2)
def train_MNL(data):
for mode in modes_list:
# availability
data[mode+'_avail'] = 1
database = db.Database("MNL_SGP", data)
beta_dic = {}
variables = {}
ASC_WALK = bioexp.Beta('B___ASC___Walk',0,None,None,1) #fixed
ASC_PT = bioexp.Beta('B___ASC___PT',0,None,None,0)
ASC_RIDEHAIL = bioexp.Beta('B___ASC___RH',0,None,None,0)
ASC_AV = bioexp.Beta('B___ASC___AV',0,None,None,0)
ASC_DRIVE = bioexp.Beta('B___ASC___Drive',0,None,None,0)
for key in att:
beta_dic[key] = {}
if key != 'Walk':
for var in z_vars:
if var not in variables:
variables[var] = bioexp.Variable(var)
beta_name = 'B___' + var + '___' + key
beta_dic[key][beta_name] = bioexp.Beta(beta_name, 0, None, None, 0)
for var in att[key]:
if var not in variables:
variables[var] = bioexp.Variable(var)
beta_name = 'B___' + var + '___' + key
beta_dic[key][beta_name] = bioexp.Beta(beta_name, 0, None, None, 0)
V = {key_choice_index['Walk']:ASC_WALK, key_choice_index['PT']:ASC_PT,
key_choice_index['RH']:ASC_RIDEHAIL,key_choice_index['AV']:ASC_AV,
key_choice_index['Drive']:ASC_DRIVE}
AV = {}
for key in att:
AV[key_choice_index[key]] = bioexp.Variable(key+'_avail')
if key != 'Walk':
for var in z_vars:
beta_name = 'B___' + var + '___' + key
V[key_choice_index[key]] += variables[var] * beta_dic[key][beta_name]
for var in att[key]:
beta_name = 'B___' + var + '___' + key
V[key_choice_index[key]] += variables[var] * beta_dic[key][beta_name]
CHOICE = bioexp.Variable('choice')
logprob = bioexp.bioLogLogit(V, AV, CHOICE)
formulas = {'loglike': logprob}
biogeme = bio.BIOGEME(database, formulas,numberOfThreads = 4)
biogeme.modelName = "MNL_SGP"
results = biogeme.estimate()
os.remove("MNL_SGP.html")
os.remove("MNL_SGP.pickle")
# Print the estimated values
betas = results.getBetaValues()
beta={}
for k, v in betas.items():
beta[k] = v
return beta
def testEstimation(self):
for k,f in self.models.items():
logging.info("Estimate {}".format(k)) ;
biogeme = bio.BIOGEME(f[0],f[1],seed=10,numberOfDraws=5)
biogeme.modelName = k
biogeme.generateHtml = False
results = biogeme.estimate()
with self.subTest(msg="{}: check final log likelihhood".format(k)):
self.assertAlmostEqual(results.data.logLike,f[2],2)
def build(self):
exclude = ((self.PURPOSE != 1) * (self.PURPOSE != 3) +
(self.CHOICE == 0)) > 0
self.database.remove(exclude)
ASC_CAR = Beta('ASC_CAR', 0, None, None, 0, 'Car cte.')
ASC_TRAIN = Beta('ASC_TRAIN', 0, None, None, 0, 'Train cte.')
ASC_SM = Beta('ASC_SM', 0, None, None, 1, 'Swissmetro cte.')
B_TIME = Beta('B_TIME', 0, None, None, 0, 'Travel time')
B_COST = Beta('B_COST', 0, None, None, 0, 'Travel cost')
SM_COST = self.SM_CO * (self.GA == 0)
TRAIN_COST = self.TRAIN_CO * (self.GA == 0)
TRAIN_TT_SCALED = DefineVariable('TRAIN_TT_SCALED', \
self.TRAIN_TT / 100.0, self.database)
TRAIN_COST_SCALED = DefineVariable('TRAIN_COST_SCALED', \
TRAIN_COST / 100, self.database)
SM_TT_SCALED = DefineVariable('SM_TT_SCALED', self.SM_TT / 100.0,
self.database)
SM_COST_SCALED = DefineVariable('SM_COST_SCALED', SM_COST / 100,
self.database)
CAR_TT_SCALED = DefineVariable('CAR_TT_SCALED', self.CAR_TT / 100,
self.database)
CAR_CO_SCALED = DefineVariable('CAR_CO_SCALED', self.CAR_CO / 100,
self.database)
self.x0 = np.zeros(4)
V1 = ASC_TRAIN + \
B_TIME * TRAIN_TT_SCALED + \
B_COST * TRAIN_COST_SCALED
V2 = ASC_SM + \
B_TIME * SM_TT_SCALED + \
B_COST * SM_COST_SCALED
V3 = ASC_CAR + \
B_TIME * CAR_TT_SCALED + \
B_COST * CAR_CO_SCALED
# Associate utility functions with the numbering of alternatives
self.V = {1: V1, 2: V2, 3: V3}
# Associate the availability conditions with the alternatives
CAR_AV_SP = DefineVariable('CAR_AV_SP', self.CAR_AV * (self.SP != 0),
self.database)
TRAIN_AV_SP = DefineVariable('TRAIN_AV_SP',
self.TRAIN_AV * (self.SP != 0),
self.database)
self.av = {1: TRAIN_AV_SP, 2: self.SM_AV, 3: CAR_AV_SP}
logprob = bioLogLogit(self.V, self.av, self.CHOICE)
self.biogeme = bio.BIOGEME(self.database, logprob)
self.biogeme.modelName = "01logit"
self.biogeme.generateHtml = False
def probas(self, betas=None):
# The choice model is a logit, with availability conditions
prob1 = models.logit(self.V, self.av, 1)
prob2 = models.logit(self.V, self.av, 2)
prob3 = models.logit(self.V, self.av, 3)
simulate = {'Train': prob1, 'SM': prob2, 'Car': prob3}
biogeme = bio.BIOGEME(self.database, simulate)
biogeme.modelName = "01logit_simul"
self.biogeme.generateHtml = False
return biogeme.simulate(betas)
def build(self):
# Parameters to be estimated
ASC_WALKING = Beta('ASC_WALKING', 0, -10, 10, 1)
ASC_CYCLING = Beta('ASC_CYCLING', 0, -10, 10, 0)
ASC_PT = Beta('ASC_PT', 0, -10, 10, 0)
ASC_DRIVING = Beta('ASC_DRIVING', 0, -10, 10, 0)
B_TIME_WALKING = Beta('B_TIME_WALKING', 0, -10, 10, 0)
B_TIME_CYCLING = Beta('B_TIME_CYCLING', 0, -10, 10, 0)
B_TIME_DRIVING = Beta('B_TIME_DRIVING', 0, -10, 10, 0)
B_COST_DRIVING = Beta('B_COST_DRIVING', 0, -10, 10, 0)
B_COST_PT = Beta('B_COST_PT', 0, -10, 10, 0)
B_TIME_PT_BUS = Beta('B_TIME_PT_BUS', 0, -10, 10, 0)
B_TIME_PT_RAIL = Beta('B_TIME_PT_RAIL', 0, -10, 10, 0)
B_TIME_PT_ACCESS = Beta('B_TIME_PT_ACCESS', 0, -10, 10, 0)
# B_INT_WALK = Beta('B_INT_WALK',0,-10,10,0)
B_TIME_PT_INT_WAIT = Beta('B_TIME_PT_INT_WAIT', 0, -10, 10, 0)
B_TRAFFIC_DRIVING = Beta('B_TRAFFIC_DRIVING', 0, -10, 10, 0)
# Utility functions
V1 = (ASC_WALKING + B_TIME_WALKING * self.dur_walking)
V2 = (ASC_CYCLING + B_TIME_CYCLING * self.dur_cycling)
V3 = (ASC_PT + B_COST_PT * self.cost_transit +
B_TIME_PT_ACCESS * self.dur_access +
B_TIME_PT_RAIL * self.dur_pt_rail +
B_TIME_PT_BUS * self.dur_pt_bus +
B_TIME_PT_INT_WAIT * self.dur_interchange_waiting)
V4 = (ASC_DRIVING + B_TIME_DRIVING * self.dur_driving +
B_COST_DRIVING * (self.cost_driving + self.con_charge) +
B_TRAFFIC_DRIVING * self.traffic_percent)
# Associate utility functions with the numbering of alternatives
V = {1: V1, 2: V2, 3: V3, 4: V4}
av = {1: 1, 2: 1, 3: 1, 4: 1}
# The choice model is a logit, with availability conditions
logprob = bioLogLogit(V, av, self.tmode)
self.biogeme = bio.BIOGEME(self.database, logprob)
self.biogeme.modelName = "LPMC_MNL_DC"
self.biogeme.generateHtml = False
self.x0 = self.biogeme.betaInitValues
def setUp(self):
np.random.seed(90267)
rnd.seed(90267)
Choice = Variable('Choice')
Variable1 = Variable('Variable1')
Variable2 = Variable('Variable2')
beta1 = Beta('beta1', 0, None, None, 0)
beta2 = Beta('beta2', 0, None, None, 0)
V1 = beta1 * Variable1
V2 = beta2 * Variable2
V3 = 0
V = {1: V1, 2: V2, 3: V3}
likelihood = models.loglogit(V, av=None, i=Choice)
self.myBiogeme = bio.BIOGEME(myData1, likelihood)
self.myBiogeme.modelName = 'simpleExample'
self.theFunction = rosenbrock()
def setUp(self):
np.random.seed(90267)
rnd.seed(90267)
Variable1 = Variable('Variable1')
Variable2 = Variable('Variable2')
beta1 = Beta('beta1', -1.0, -3, 3, 0)
beta2 = Beta('beta2', 2.0, -3, 10, 0)
likelihood = -beta1**2 * Variable1 - exp(
beta2 * beta1) * Variable2 - beta2**4
simul = beta1 / Variable1 + beta2 / Variable2
dictOfExpressions = {
'loglike': likelihood,
'beta1': beta1,
'simul': simul
}
self.myBiogeme = bio.BIOGEME(myData1, dictOfExpressions)
self.myBiogeme.modelName = 'simpleExample'
def predict_NL_2(betas, biogeme_file, data):
for mode in modes_list:
# availability
data[mode+'_avail'] = 1
database = db.Database("NL_SGP", data)
# The choice model is a nested logit
prob_Walk = biomodels.nested(biogeme_file['V'], biogeme_file['av'], biogeme_file['nests'], key_choice_index['Walk'])
prob_PT = biomodels.nested(biogeme_file['V'], biogeme_file['av'], biogeme_file['nests'], key_choice_index['PT'])
prob_RH = biomodels.nested(biogeme_file['V'], biogeme_file['av'], biogeme_file['nests'], key_choice_index['RH'])
prob_AV = biomodels.nested(biogeme_file['V'], biogeme_file['av'], biogeme_file['nests'], key_choice_index['AV'])
prob_Drive = biomodels.nested(biogeme_file['V'], biogeme_file['av'], biogeme_file['nests'], key_choice_index['Drive'])
simulate = {'prob_Walk': prob_Walk,
'prob_PT': prob_PT,
'prob_RH': prob_RH,
'prob_AV':prob_AV,
'prob_Drive':prob_Drive}
biogeme = bio.BIOGEME(database, simulate)
# Extract the values that are necessary
betaValues = betas
# simulatedValues is a Panda dataframe with the same number of rows as
# the database, and as many columns as formulas to simulate.
simulatedValues = biogeme.simulate(betaValues)
prob_list = list(simulatedValues.columns)
data_test = data
for key in prob_list:
data_test[key] = 0
data_test.loc[:,prob_list] = simulatedValues.loc[:, prob_list]
data_test['max_prob'] = data_test[prob_list].max(axis=1)
data_test['CHOOSE'] = 0
for mode in key_choice_index:
col_nameprob = 'prob_' + mode
data_test.loc[data_test[col_nameprob]==data_test['max_prob'],'CHOOSE'] = key_choice_index[mode]
acc = len(data_test.loc[data_test['CHOOSE']==data_test['choice']])/len(data_test)
return acc, data_test
def train_MNL(data):
for mode in modes_list:
# availability
data[mode+'_avail'] = 1
database = db.Database("MNL_Train", data)
beta_dic = {}
variables = {}
ASC_1 = bioexp.Beta('B___ASC___choice1',0,None,None,1) #fixed
ASC_2 = bioexp.Beta('B___ASC___choice2',0,None,None,0)
for key in att:
beta_dic[key] = {}
for var in att[key]:
if var not in variables:
variables[var] = bioexp.Variable(var)
beta_name = 'B___' + var + '___' + key
beta_dic[key][beta_name] = bioexp.Beta(beta_name, 0, None, None, 0)
V = {key_choice_index['choice1']:ASC_1, key_choice_index['choice2']:ASC_2}
AV = {}
for key in att:
AV[key_choice_index[key]] = bioexp.Variable(key+'_avail')
for var in att[key]:
beta_name = 'B___' + var + '___' + key
V[key_choice_index[key]] += variables[var] * beta_dic[key][beta_name]
CHOICE = bioexp.Variable('choice')
logprob = bioexp.bioLogLogit(V, AV, CHOICE)
formulas = {'loglike': logprob}
biogeme = bio.BIOGEME(database, formulas,numberOfThreads = 4)
biogeme.modelName = "MNL_Train"
results = biogeme.estimate()
os.remove("MNL_Train.html")
os.remove("MNL_Train.pickle")
# Print the estimated values
betas = results.getBetaValues()
beta={}
for k, v in betas.items():
beta[k] = v
return beta
def testEstimation(self):
AutoTime = Variable('AutoTime')
TransitTime = Variable('TransitTime')
Choice = Variable('Choice')
ASC_CAR = Beta('ASC_CAR', 0, None, None, 0)
B_TIME = Beta('B_TIME', 0, None, None, 0)
V = {0: ASC_CAR + B_TIME * AutoTime, 1: B_TIME * TransitTime}
av = {0: 1, 1: 1}
logprob = models.loglogit(V, av, Choice)
biogeme = bio.BIOGEME(self.database, logprob)
biogeme.modelName = 'test'
biogeme.generateHtml = False
results = biogeme.estimate()
self.assertAlmostEqual(results.data.logLike, -6.166042, 2)
B_COST * SM_COST_SCALED
V3 = ASC_CAR + \
B_TIME * models.boxcox(CAR_TT_SCALED, LAMBDA) + \
B_COST * CAR_CO_SCALED
# Associate utility functions with the numbering of alternatives
V = {1: V1, 2: V2, 3: V3}
# Associate the availability conditions with the alternatives
av = {1: TRAIN_AV_SP, 2: SM_AV, 3: CAR_AV_SP}
# Definition of the model. This is the contribution of each
# observation to the log likelihood function.
logprob = models.loglogit(V, av, CHOICE)
# Define level of verbosity
logger = msg.bioMessage()
logger.setSilent()
#logger.setWarning()
#logger.setGeneral()
#logger.setDetailed()
# Create the Biogeme object
biogeme = bio.BIOGEME(database, logprob)
biogeme.modelName = '08boxcox'
# Estimate the parameters
results = biogeme.estimate()
pandasResults = results.getEstimatedParameters()
print(pandasResults)
def testEstimation(self):
biogeme = bio.BIOGEME(database, logprob, numberOfDraws=5, seed=10)
results = biogeme.estimate()
self.assertAlmostEqual(results.data.logLike, -4705.638407401872, 2)
B_COST = Beta('B_COST', 0, None, None, 0)
tau1 = Beta('tau1', -1, None, 0, 0)
delta2 = Beta('delta2', 2, 0, None, 0)
tau2 = tau1 + delta2
TRAIN_COST = TRAIN_CO * (GA == 0)
TRAIN_TT_SCALED = DefineVariable('TRAIN_TT_SCALED',\
TRAIN_TT / 100.0,database)
TRAIN_COST_SCALED = DefineVariable('TRAIN_COST_SCALED',\
TRAIN_COST / 100,database)
# Utility
U = B_TIME * TRAIN_TT_SCALED + B_COST * TRAIN_COST_SCALED
ChoiceProba = {
1: 1 - dist.logisticcdf(U - tau1),
2: dist.logisticcdf(U - tau1) - dist.logisticcdf(U - tau2),
3: dist.logisticcdf(U - tau2)
}
logprob = log(Elem(ChoiceProba, CHOICE))
biogeme = bio.BIOGEME(database, logprob, numberOfThreads=1)
biogeme.modelName = "18ordinalLogit"
results = biogeme.estimate()
print("Results=", results)
V2 = ASC_SM + \
B_TIME_RND * SM_TT_SCALED + \
B_COST * SM_COST_SCALED
V3 = ASC_CAR + \
B_TIME_RND * CAR_TT_SCALED + \
B_COST * CAR_CO_SCALED
# Associate utility functions with the numbering of alternatives
V = {1: V1,
2: V2,
3: V3}
# Associate the availability conditions with the alternatives
av = {1: TRAIN_AV_SP,
2: SM_AV,
3: CAR_AV_SP}
# The choice model is a logit, with availability conditions
prob = models.logit(V, av, CHOICE)
logprob = log(MonteCarlo(prob))
R = 2000
biogeme = bio.BIOGEME(database, logprob, numberOfDraws=R)
biogeme.modelName = '07estimationMonteCarlo_mlhs'
results = biogeme.estimate()
# Get the results in a pandas table
pandasResults = results.getEstimatedParameters()
print(pandasResults)
condlike = P_Envir01 * \
P_Envir02 * \
P_Envir03 * \
P_Mobil11 * \
P_Mobil14 * \
P_Mobil16 * \
P_Mobil17 * \
condprob
# We integrate over omega using Monte-Carlo integration
loglike = log(MonteCarlo(condlike))
# Define level of verbosity
logger = msg.bioMessage()
#logger.setSilent()
#logger.setWarning()
logger.setGeneral()
#logger.setDetailed()
# Create the Biogeme object
biogeme = bio.BIOGEME(database, loglike, numberOfDraws=20000)
biogeme.modelName = '06serialCorrelation'
# Estimate the parameters
results = biogeme.estimate()
print(f'Estimated betas: {len(results.data.betaValues)}')
print(f'Final log likelihood: {results.data.logLike:.3f}')
print(f'Output file: {results.data.htmlFileName}')
results.writeLaTeX()
print(f'LaTeX file: {results.data.latexFileName}')
def testEstimation(self):
biogeme = bio.BIOGEME(database, logprob)
results = biogeme.estimate()
self.assertAlmostEqual(results.data.logLike, -5215.072, 2)
# Define level of verbosity
logger = msg.bioMessage()
#logger.setSilent()
#logger.setWarning()
logger.setGeneral()
#logger.setDetailed()
# These notes will be included as such in the report file.
userNotes = ('Example of a logit model with three alternatives: Train, Car and Swissmetro.'
' Same as 01logit, using bioLinearUtility, and introducing some options '
'and features.')
# Create the Biogeme object
biogeme = bio.BIOGEME(database, logprob, numberOfThreads=2, userNotes=userNotes)
biogeme.modelName = '01logitBis'
# Estimate the parameters
results = biogeme.estimate(bootstrap=100,
algorithm=opt.bioNewton,
algoParameters={'hamabs': True},
saveIterations=True)
biogeme.createLogFile(verbosity=3)
# Get the results in a pandas table
print('Parameters')
print('----------')
pandasResults = results.getEstimatedParameters()
print(pandasResults)
def testEstimation(self):
biogeme = bio.BIOGEME(database, logprob, numberOfDraws=5, seed=10)
results = biogeme.estimate()
self.assertAlmostEqual(results.data.logLike, -4557.1998156795935, 2)
CAR_CO_SCALED = DefineVariable('CAR_CO_SCALED', CAR_CO / 100, database)
V1 = ASC_TRAIN + \
B_TIME * TRAIN_TT_SCALED + \
B_COST * TRAIN_COST_SCALED
V2 = ASC_SM + \
B_TIME * SM_TT_SCALED + \
B_COST * SM_COST_SCALED
V3 = ASC_CAR + \
B_TIME * CAR_TT_SCALED + \
B_COST * CAR_CO_SCALED
# Associate utility functions with the numbering of alternatives
V = {1: V1, 2: V2, 3: V3}
# Associate the availability conditions with the alternatives
CAR_AV_SP = DefineVariable('CAR_AV_SP', CAR_AV * (SP != 0), database)
TRAIN_AV_SP = DefineVariable('TRAIN_AV_SP', TRAIN_AV * (SP != 0), database)
av = {1: TRAIN_AV_SP, 2: SM_AV, 3: CAR_AV_SP}
logprob = bioLogLogit(V, av, CHOICE)
weight = 8.890991e-01 * (1.0 * (GROUP == 2) + 1.2 * (GROUP == 3))
formulas = {'loglike': logprob, 'weight': weight}
biogeme = bio.BIOGEME(database, formulas)
biogeme.modelName = "02weight"
results = biogeme.estimate()
print("Results=", results)
def testEstimation(self):
logprob = models.loglogit(V, av, CHOICE)
biogeme = bio.BIOGEME(database, logprob)
results = biogeme.estimate()
self.assertAlmostEqual(results.data.logLike, -5423.299, 2)
CAR_TT_SCALED = CAR_TT / 100
CAR_CO_SCALED = CAR_CO / 100
# Definition of the utility functions
V1 = ASC_TRAIN + \
B_TIME_RND * TRAIN_TT_SCALED + \
B_COST * TRAIN_COST_SCALED
V2 = ASC_SM + \
B_TIME_RND * SM_TT_SCALED + \
B_COST * SM_COST_SCALED
V3 = ASC_CAR + \
B_TIME_RND * CAR_TT_SCALED + \
B_COST * CAR_CO_SCALED
# Associate utility functions with the numbering of alternatives
V = {1: V1, 2: V2, 3: V3}
# Associate the availability conditions with the alternatives
av = {1: TRAIN_AV_SP, 2: SM_AV, 3: CAR_AV_SP}
# The choice model is a logit, with availability conditions
integrand = models.logit(V, av, CHOICE)
numericalI = Integrate(integrand * density, 'omega')
simulate = {'Numerical': numericalI}
biogeme = bio.BIOGEME(database, simulate)
results = biogeme.simulate()
print('Mixture of logit - numerical integration: ',
results.iloc[0]['Numerical'])