def loocv(d, formula, output):
print('processing loocv', file=file)
print('processing loocv')
loo = cross_val.LeaveOneOut(len(d.index))
error_sum = 0
for train_index, test_index in loo:
# print ("TRAIN:", train_index, "TEST:", test_index)
a_train, a_test = cross_val.split(train_index, test_index, d)
d_train = pd.DataFrame(a_train, columns=d.columns)
d_test = pd.DataFrame(a_test, columns=d.columns)
for x in d.columns:
d_train[x] = d_train[x].astype(d[x].dtypes.name)
d_test[x] = d_test[x].astype(d[x].dtypes.name)
nuc = sm.ols(formula, data=d_train).fit()
y = nuc.predict(d_test)
error_sum += (y[0] - d_test['ReservesLevel'][0])**2
print("loocv MSE= ", (error_sum / len(d.index)), file=file)
print("loocv MSE= ", (error_sum / len(d.index)))
from statsmodels.sandbox.tools import cross_val
if __name__ == '__main__':
#A: josef-pktd
import statsmodels.api as sm
#from statsmodels.datasets.longley import load
from statsmodels.datasets.stackloss import load
data = load()
data.exog = sm.tools.add_constant(data.exog, prepend=False)
resols = sm.OLS(data.endog, data.exog).fit()
print('\n OLS leave 1 out')
for inidx, outidx in cross_val.LeaveOneOut(len(data.endog)):
res = sm.OLS(data.endog[inidx], data.exog[inidx, :]).fit()
print(data.endog[outidx],
res.model.predict(res.params, data.exog[outidx, :], end=' '))
print(data.endog[outidx] -
res.model.predict(res.params, data.exog[outidx, :]))
print('\n OLS leave 2 out')
resparams = []
for inidx, outidx in cross_val.LeavePOut(len(data.endog), 2):
res = sm.OLS(data.endog[inidx], data.exog[inidx, :]).fit()
#print data.endog[outidx], res.model.predict(data.exog[outidx,:]),
#print ((data.endog[outidx] - res.model.predict(data.exog[outidx,:]))**2).sum()
resparams.append(res.params)
resparams = np.array(resparams)
import pandas as pd
import statsmodels.formula.api as sm
import statsmodels.api as sma
import statsmodels.sandbox.tools.cross_val as cross_val
d = pd.read_csv("auto.csv")
loo = cross_val.LeaveOneOut(len(d.index))
error_sum = 0
for train_index, test_index in loo:
# print ("TRAIN:", train_index, "TEST:", test_index)
a_train, a_test = cross_val.split(train_index, test_index, d)
d_train = pd.DataFrame(a_train, columns=d.columns)
d_test = pd.DataFrame(a_test, columns=d.columns)
for x in d.columns:
d_train[x] = d_train[x].astype(d[x].dtypes.name)
d_test[x] = d_test[x].astype(d[x].dtypes.name)
nuc = sm.ols("mpg~horsepower", data=d_train).fit()
y = nuc.predict(d_test)
error_sum += (y[0] - d_test["mpg"][0])**2
print("MSE= ", (error_sum / len(d.index)))