Python Statadata Exemples

Exemple #1

Fichier : econometrics_regression_adjustment.py Projet : NeckiiZ/workrobot

# coding = UTF-8

# ===================================
# 这个程序用来实现Regression-Adjustment
# ===================================

import numpy as np
from libs.imexport.class_statadata import Statadata
from libs.statmodel.class_lm import Lm
from rpy2.robjects.packages import importr
from libs.embeddedR.class_renv import REnv

# 1. 导入数据
stata_file = r'd:\data\userdata\jtrain2.dta'
stdata = Statadata(stata_file)
raw_data = stdata.read()

# 2. 比较两组对象
grouped = raw_data.groupby('train')
compared = grouped.aggregate(np.mean)
print(compared.T)

for _, group in grouped:
    print(group.mean())

# 3. 回归
reg_obj = Lm(data=raw_data, formula='re78 ~ train')
call_obj = reg_obj()
print(reg_obj.coefs)

reg_obj = Lm(

Exemple #2

Fichier : class_bootstrap.py Projet : plutoese/workrobot

        bootBias = pd.DataFrame(boot_data-original).mean()

        pdict = OrderedDict()
        pdict['original'] = original
        pdict['bootSE'] = bootSE
        pdict['bootBias'] = bootBias
        return {'boot_data':boot_data, 'stat':pd.DataFrame(pdict)}


if __name__ == '__main__':
    '''
    stata_file = 'd:/data/userdata/newcity.dta'
    stdata = Statadata(stata_file)
    city_data = stdata.read()

    boot = Bootstrap(data=city_data, func='function(d, w) sum(d$x * w)/sum(d$u * w)')
    result = boot.run(stype='"w"')

    print(result['stat'])

    '''
    stata_file = 'd:/data/userdata/mtcars.dta'
    stdata = Statadata(stata_file)
    cars_data = stdata.read()

    boot = Bootstrap(data=cars_data, func='function(formula, data, indices){\n\td <- data[indices,]\n\tfit <- lm(formula, data=d)\n\treturn(coef(fit))}')
    result = boot.run(formula='mpg~wt+disp')

    print(result['stat'])

Exemple #3

    def adj_Rsquared(self):
        return self.robj.get_var('slm_obj$adj.r.squared')

    @property
    def cov(self):
        return self.robj.get_var('slm_obj$cov.unscaled')

    @property
    def qr(self):
        return self.robj.get_var('lm_obj$qr$qr')

if __name__ == '__main__':
    rthread = RThread()

    stata_file = r'D:\data\test\wage1.dta'
    stdata = Statadata(stata_file)
    mdata = stdata.read()
    rdata = mdata[['lwage','educ','exper','tenure']]

    rthread.create_vars(rdata)
    print(rthread.get_var('lwage'))

    #lm = Rreg(robj=rthread,xvar=['educ','exper','tenure'],yvar='lwage')
    '''
    print(lm.coefficients)
    print(lm.fstatistic)
    print(lm.sigma)
    print(lm.df)
    print(lm.qr)
    print(lm.Rsquared)
    print(lm.adj_Rsquared)

Exemple #4

Fichier : econometrics_regression_adjustment.py Projet : plutoese/workrobot

# coding = UTF-8

# ===================================
# 这个程序用来实现Regression-Adjustment
# ===================================

import numpy as np
from libs.imexport.class_statadata import Statadata
from libs.statmodel.class_lm import Lm
from rpy2.robjects.packages import importr
from libs.embeddedR.class_renv import REnv

# 1. 导入数据
stata_file = r'd:\data\userdata\jtrain2.dta'
stdata = Statadata(stata_file)
raw_data = stdata.read()

# 2. 比较两组对象
grouped = raw_data.groupby('train')
compared = grouped.aggregate(np.mean)
print(compared.T)

for _, group in grouped:
    print(group.mean())

# 3. 回归
reg_obj = Lm(data=raw_data, formula = 're78 ~ train')
call_obj = reg_obj()
print(reg_obj.coefs)

reg_obj = Lm(data=raw_data, formula = 're78 ~ train + re74 + re75 + age + agesq + nodegree + married + black + hisp')