for year in article.GSSYearsUsed:
design = dataCont.df.loc[year, [DV] + article.IVs]
# design = design.fillna(design.mean())
formula = GU.createFormula(dataCont, design)
# results = smf.ols(formula2, data=design.dropna()).fit()
if len(design.dropna()) <= design.shape[1]:
nominals = GU.createFormula(dataCont, design, return_nominals=True)
non_nominals = list(set(design.columns) - set(nominals)) # list because sets are unhashable and cant be used for indices
if len(non_nominals)>0:
design[non_nominals] = design[non_nominals].fillna(design[non_nominals].mean()) # the naive way
if len(nominals)>0:
design[nominals] = design[nominals].fillna(design[nominals].mode())
design = GU.removeConstantColumns(design.dropna())
print smf.ols(formula, data=design.dropna()).fit().summary()
break
# In[49]:
print formula
design.head()
# In[68]:
DV = 'GRASS'
for year in article.GSSYearsUsed:
design = dataCont.df.loc[year, [DV] + article.IVs]
