Data Loading¶

import nltk
nltk.download('stopwords')

[nltk_data] Downloading package stopwords to /root/nltk_data...
[nltk_data]   Package stopwords is already up-to-date!

True

from sklearn.metrics import accuracy_score

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split as tt
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.naive_bayes import BernoulliNB

from sklearn.naive_bayes import MultinomialNB
from sklearn.svm import LinearSVC
from sklearn.linear_model import LogisticRegression
from sklearn.pipeline import Pipeline
from sklearn.feature_extraction.text import TfidfVectorizer
from nltk.corpus import stopwords
from sklearn.multiclass import OneVsRestClassifier

stop_words = set(stopwords.words('english'))

from collections import OrderedDict

df = pd.read_csv('Training_Data.csv',encoding = "ISO-8859-1")

test = df[df['Growth']== 0][df['Inflation']==0][df['Cad']==0][df['Money']==0][df['Fiscal']==0][df['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

test.describe()

df1 = pd.concat([test,df])

train = df1.drop_duplicates(keep=False)

train.describe()

train['Sentence']

7         In addition to weak demand, the sharp fall in ...
9         There could be improvement in the days to com...
45        The chart shows that average inflation expecta...
47        add_main_imageThe years of low inflation from ...
48        Conversely, the years of high inflation betwee...
                                ...                        
220840    The future of cryptocurrency startups is uncer...
220841    A draft bill, Banning of Crypto Currency and R...
220843    Also, billions of people worldwide are introd...
220847    A person aware of the developments had told ET...
220848    Sathvik Vishwanath, CEO and cofounder, Unocoin...
Name: Sentence, Length: 31359, dtype: object

NLP MultiClass¶

NB_pipeline = Pipeline([
                ('tfidf', TfidfVectorizer(stop_words=stop_words)),
                ('clf', OneVsRestClassifier(MultinomialNB(
                    fit_prior=True, class_prior=None))),
            ])

 categories = ['Growth', 'Inflation', 'Cad', 'Money','Fiscal', 'Other']

x_train,x_test = tt(train,test_size = 0.5,random_state = 1)

for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    NB_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = NB_pipeline.predict(x_test['Sentence'])
    print('Test accuracy is {}'.format(accuracy_score(x_test[category], prediction)))

... Processing Growth
Test accuracy is 0.9246173469387755
... Processing Inflation
Test accuracy is 0.9117984693877551
... Processing Cad
Test accuracy is 0.8779336734693878
... Processing Money
Test accuracy is 0.9185586734693878
... Processing Fiscal
Test accuracy is 0.9452168367346939
... Processing Other
Test accuracy is 0.9584183673469387

SVC_pipeline = Pipeline([
                ('tfidf', TfidfVectorizer(stop_words=stop_words)),
                ('clf', OneVsRestClassifier(LinearSVC(), n_jobs=1)),
            ])

for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    SVC_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = SVC_pipeline.predict(x_test['Sentence'])
    print('Test accuracy is {}'.format(accuracy_score(x_test[category], prediction)))

... Processing Growth
Test accuracy is 0.9815051020408163
... Processing Inflation
Test accuracy is 0.9830357142857142
... Processing Cad
Test accuracy is 0.9827168367346939
... Processing Money
Test accuracy is 0.9900510204081633
... Processing Fiscal
Test accuracy is 0.99375
... Processing Other
Test accuracy is 0.9926020408163265

LogReg_pipeline = Pipeline([
                ('tfidf', TfidfVectorizer(stop_words=stop_words)),
                ('clf', OneVsRestClassifier(LogisticRegression(solver='sag'), n_jobs=1)),
            ])

for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    LogReg_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = LogReg_pipeline.predict(x_test['Sentence'])
    print('Test accuracy is {}'.format(accuracy_score(x_test[category], prediction)))

... Processing Growth
Test accuracy is 0.973405612244898
... Processing Inflation
Test accuracy is 0.9677295918367347
... Processing Cad
Test accuracy is 0.963265306122449
... Processing Money
Test accuracy is 0.9741709183673469
... Processing Fiscal
Test accuracy is 0.9841198979591836
... Processing Other
Test accuracy is 0.9791454081632653

Test Dataset NLP Multiclass model¶

x_train,x_test = tt(train,test_size = 0.5,random_state = 9)

for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    SVC_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = SVC_pipeline.predict(test['Sentence'])
    test[category] = prediction

... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other

anomoly = test[test['Growth']== 0][test['Inflation']==0][test['Cad']==0][test['Money']==0][test['Fiscal']==0][test['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

anomoly.describe()

x_train,x_test = tt(train,test_size = 0.5,random_state = 8)

train1 = pd.concat([test,train,anomoly])

train = train1.drop_duplicates(keep=False)

train.describe()

for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    LogReg_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = LogReg_pipeline.predict(anomoly['Sentence'])
    anomoly[category] = prediction

... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other

Check = anomoly[anomoly['Growth']== 0][anomoly['Inflation']==0][anomoly['Cad']==0][anomoly['Money']==0][anomoly['Fiscal']==0][anomoly['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

Check.describe()

train1 = pd.concat([Check,train,anomoly])

train = train1.drop_duplicates(keep=False)

x_train,x_test = tt(train,test_size = 0.5,random_state = 5)

for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    SVC_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = SVC_pipeline.predict(Check['Sentence'])
    Check[category] = prediction

... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other

Check1 = Check[Check['Growth']== 0][Check['Inflation']==0][Check['Cad']==0][Check['Money']==0][Check['Fiscal']==0][Check['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

Check1.describe()

train1 = pd.concat([Check,train,Check1])
train = train1.drop_duplicates(keep=False)

train.describe()

train.describe()

x_train,x_test = tt(train,test_size = 0.5,random_state = 4)

for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    SVC_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = SVC_pipeline.predict(Check1['Sentence'])
    Check1[category] = prediction

... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other

Check11 = Check1[Check1['Growth']== 0][Check1['Inflation']==0][Check1['Cad']==0][Check1['Money']==0][Check1['Fiscal']==0][Check1['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

Check11.describe()

train1 = pd.concat([Check11,train,Check1])
train = train1.drop_duplicates(keep=False)

x_train,x_test = tt(train,test_size = 0.5,random_state = 4)

x_train,x_test = tt(train,test_size = 0.5,random_state = 4)
for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    SVC_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = SVC_pipeline.predict(Check11['Sentence'])
    Check11[category] = prediction

... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other

Check111 = Check11[Check11['Growth']== 0][Check11['Inflation']==0][Check11['Cad']==0][Check11['Money']==0][Check11['Fiscal']==0][Check11['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

Check111.describe()

train1 = pd.concat([Check111,train,Check11])
train = train1.drop_duplicates(keep=False)

x_train,x_test = tt(train,test_size = 0.5,random_state = 4)
for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    SVC_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = SVC_pipeline.predict(Check111['Sentence'])
    Check111[category] = prediction

... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other

Check1111 = Check111[Check111['Growth']== 0][Check111['Inflation']==0][Check111['Cad']==0][Check111['Money']==0][Check111['Fiscal']==0][Check111['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

Check1111.describe()

train1 = pd.concat([Check1111,train,Check111])
train = train1.drop_duplicates(keep=False)

x_train,x_test = tt(train,test_size = 0.4,random_state = 5)
for category in categories:
    print('... Processing {}'.format(category))
    # train the model using X_dtm & y
    SVC_pipeline.fit(x_train['Sentence'], x_train[category])
    # compute the testing accuracy
    prediction = SVC_pipeline.predict(Check1111['Sentence'])
    Check1111[category] = prediction

... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other

Check11111 = Check1111[Check1111['Growth']== 0][Check1111['Inflation']==0][Check1111['Cad']==0][Check1111['Money']==0][Check1111['Fiscal']==0][Check1111['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

Check11111.describe()

train1 = pd.concat([Check11111,train,Check1111])
train = train1.drop_duplicates(keep=False)

train.to_csv("Out.csv")

checking¶

test = Check11111

for i in range(10):
    print(i)
    x_train,x_test = tt(train,test_size = 0.5,random_state = i)
    for category in categories:
        print('... Processing {}'.format(category))
        # train the model using X_dtm & y
        SVC_pipeline.fit(x_train['Sentence'], x_train['Growth'])
        # compute the testing accuracy
        prediction = SVC_pipeline.predict(test['Sentence'])
        test[category] = prediction
    anomoly = test[test['Growth']== 0][test['Inflation']==0][test['Cad']==0][test['Money']==0][test['Fiscal']==0][test['Other']==0]
    test = anomoly
    train1 = pd.concat([test,train,anomoly])
    train = train1.drop_duplicates(keep=False)

0
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
1
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
2
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
3
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
4
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
5
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
6
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
7
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
8
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other
9
... Processing Growth
... Processing Inflation
... Processing Cad
... Processing Money
... Processing Fiscal
... Processing Other

train.describe()

test.describe()

train2 = pd.concat([df,train])

train2 = train2.drop_duplicates(keep=False)

train2.describe()

train = pd.concat([train,anomoly])

acc = {}
for i in range(2):
  print(i)
  x_train,x_test = tt(train,test_size = 0.5,random_state = i)
  for category in categories:
      print('... Processing {}'.format(category))
      # train the model using X_dtm & y
      SVC_pipeline.fit(x_train['Sentence'], x_train['category'])
      # compute the testing accuracy

      # test[category] = prediction
  # anomoly = test[test['Growth']== 0][test['Inflation']==0][test['Cad']==0][test['Money']==0][test['Fiscal']==0][test['Other']==0]
  # test = anomoly
  # train1 = pd.concat([test,train,anomoly])
  # train = train1.drop_duplicates(keep=False)

0
1

acc

{1: 'epoch', 'Score': 0.9539698868418591}

NLP Model¶

vect = CountVectorizer()

vect.fit(train['Sentence'])
counts= vect.transform(train['Sentence'])

tf = TfidfTransformer()
train1 = tf.fit_transform(counts)

train1.shape[1]

43322

x_train,x_test,y_train,y_test=tt(train1,np.asarray(train['Growth']),test_size = 0.3,stratify = np.asarray(train['Growth']),random_state = 1)

modl=BernoulliNB().fit(x_train,y_train)

from sklearn.metrics import accuracy_score

acc = modl.predict(x_test)

accuracy_score(acc,y_test)

0.8827938027459378

x_train,x_test,y_train,y_test=tt(train1,np.asarray(train['Inflation']),test_size = 0.3,stratify = np.asarray(train['Growth']),random_state = 1)
modl=BernoulliNB().fit(x_train,y_train)
acc = modl.predict(x_test)
accuracy_score(acc,y_test)

0.9286906411386825

x_train,x_test,y_train,y_test=tt(train1,np.asarray(train['Cad']),test_size = 0.3,stratify = np.asarray(train['Growth']),random_state = 1)
modl=BernoulliNB().fit(x_train,y_train)
acc = modl.predict(x_test)
accuracy_score(acc,y_test)

0.944435697191082

x_train,x_test,y_train,y_test=tt(train1,np.asarray(train['Money']),test_size = 0.3,stratify = np.asarray(train['Growth']),random_state = 1)
modl=BernoulliNB().fit(x_train,y_train)
acc = modl.predict(x_test)
accuracy_score(acc,y_test)

0.9635816853507998

x_train,x_test,y_train,y_test=tt(train1,np.asarray(train['Fiscal']),test_size = 0.3,stratify = np.asarray(train['Growth']),random_state = 1)
modl=BernoulliNB().fit(x_train,y_train)
acc = modl.predict(x_test)
accuracy_score(acc,y_test)

0.9852311374228492

Test dataset¶

vect1 = CountVectorizer()
vect1.fit(test['Sentence'])
counts= vect.transform(test['Sentence'])
tf1 = TfidfTransformer()
test1 = tf1.fit_transform(counts)

Growth Classification

modl=BernoulliNB().fit(train1,train['Growth'])
y = modl.predict(test1)
test['Growth'] = y

Inflation Classification

modl=BernoulliNB().fit(train1,train['Inflation'])
y = modl.predict(test1)
test['Inflation'] = y

Cad

modl=BernoulliNB().fit(train1,train['Cad'])
y = modl.predict(test1)
test['Cad'] = y

Money

modl=BernoulliNB().fit(train1,train['Money'])
y = modl.predict(test1)
test['Money'] = y

Fiscal

modl=BernoulliNB().fit(train1,train['Fiscal'])
y = modl.predict(test1)
test['Fiscal'] = y

Other

modl=BernoulliNB().fit(train1,train['Other'])
y = modl.predict(test1)
test['Other'] = y

anomoly = test[test['Growth']== 0][test['Inflation']==0][test['Cad']==0][test['Money']==0][test['Fiscal']==0][test['Other']==0]

/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.
  """Entry point for launching an IPython kernel.

t = pd.concat([anomoly,test])

test = t.drop_duplicates(keep=False)

test.describe()

vect1 = CountVectorizer()
vect1.fit(anomoly['Sentence'])
counts= vect.transform(anomoly['Sentence'])
tf1 = TfidfTransformer()
anomoly1 = tf1.fit_transform(counts)

modl=BernoulliNB().fit(train1,train['Growth'])Other = anomoly[anomoly['Growth']== 0][anomoly['Inflation']==0][anomoly['Cad']==0][anomoly['Money']==0][anomoly['Fiscal']==0][anomoly['Other']==0]
anomoly['Growth'] = modl.predict(anomoly1)

  File "<ipython-input-353-25eaa21c8070>", line 1
    modl=BernoulliNB().fit(train1,train['Growth'])Other = anomoly[anomoly['Growth']== 0][anomoly['Inflation']==0][anomoly['Cad']==0][anomoly['Money']==0][anomoly['Fiscal']==0][anomoly['Other']==0]
                                                      ^
SyntaxError: invalid syntax

modl=BernoulliNB().fit(train1,train['Inflation'])
anomoly['Inflation'] = modl.predict(anomoly1)

modl=BernoulliNB().fit(train1,train['Cad'])
anomoly['Cad'] = modl.predict(anomoly1)

modl=BernoulliNB().fit(train1,train['Money'])
anomoly['Money'] = modl.predict(anomoly1)

modl=BernoulliNB().fit(train1,train['Fiscal'])
anomoly['Fiscal'] = modl.predict(anomoly1)

modl=BernoulliNB().fit(train1,train['Other'])
anomoly['Other'] = modl.predict(anomoly1)

Other = anomoly[anomoly['Growth']== 0][anomoly['Inflation']==0][anomoly['Cad']==0][anomoly['Money']==0][anomoly['Fiscal']==0][anomoly['Other']==0]

Other

References¶

class NeuralNet():
	def __init__(self,input):
		np.random.seed(1)
		self.weights = 2 * np.random.random(input) - 1
		print("The weights are ",self.weights)
	def sigmoid(self,a):
		return 1/(1 + np.exp(-a))	

	def der_sigmoid(self,a):
		return a * (1 - a)	

	def think(self,ip):
		ip = ip.astype(float)
		output = self.sigmoid(np.dot(ip,self.weights))
		return output

	def training(self,input_layer,T_outputs,iterations):
		for i in range(iterations):	
			output = self.think(input_layer)
			error = T_outputs - output
			adjust = error*self.der_sigmoid(output)
			self.weights += np.dot(input_layer.T,adjust)

from sklearn.datasets import fetch_20newsgroups

categories = ['alt.atheism', 'soc.religion.christian','comp.graphics', 'sci.med']
news_train = fetch_20newsgroups(subset = 'train', categories= categories)
news_test = fetch_20newsgroups(subset = 'test',categories= categories)

from sklearn.pipeline import Pipeline
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.naive_bayes import MultinomialNB

text_clf = Pipeline([('vect', TfidfVectorizer()), 
                      ('clf', MultinomialNB()) ])

# train the model
text_clf.fit(news_train.data, news_train.target)
# Predict the test cases
predicted = text_clf.predict(news_test.data)

from sklearn import metrics
from sklearn.metrics import accuracy_score
import numpy as np

print('Accuracy achieved is ' + str(np.mean(predicted == news_test.target)))
print(metrics.classification_report(news_test.target, predicted, target_names=news_test.target_names)),
metrics.confusion_matrix(news_test.target, predicted)

news_train.target

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	189494.000000	189494.0	189494.0	189494.0	189494.0	189494.0	189494.0
mean	109299.531917	0.0	0.0	0.0	0.0	0.0	0.0
std	63907.299044	0.0	0.0	0.0	0.0	0.0	0.0
min	0.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	53950.250000	0.0	0.0	0.0	0.0	0.0	0.0
50%	107822.500000	0.0	0.0	0.0	0.0	0.0	0.0
75%	164733.750000	0.0	0.0	0.0	0.0	0.0	0.0
max	220852.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	31359.000000	31359.000000	31359.000000	31359.000000	31359.000000	31359.000000	31359.000000
mean	117232.943557	0.534073	0.411333	0.142957	0.096017	0.053222	0.040754
std	62395.971409	0.498846	0.492083	0.350035	0.294619	0.224480	0.197723
min	7.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	63979.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
50%	124876.000000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
75%	170273.500000	1.000000	1.000000	0.000000	0.000000	0.000000	0.000000
max	220848.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	126921.000000	126921.0	126921.0	126921.0	126921.0	126921.0	126921.0
mean	110844.316488	0.0	0.0	0.0	0.0	0.0	0.0
std	64008.795475	0.0	0.0	0.0	0.0	0.0	0.0
min	0.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	55265.000000	0.0	0.0	0.0	0.0	0.0	0.0
50%	110605.000000	0.0	0.0	0.0	0.0	0.0	0.0
75%	166511.000000	0.0	0.0	0.0	0.0	0.0	0.0
max	220852.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	93932.000000	93932.000000	93932.000000	93932.000000	93932.000000	93932.000000	93932.000000
mean	109860.770398	0.611698	0.277541	0.100413	0.075012	0.029500	0.033769
std	63406.210596	0.487367	0.447788	0.300552	0.263412	0.169204	0.180635
min	3.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	55141.750000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
50%	110184.500000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
75%	164479.250000	1.000000	1.000000	0.000000	0.000000	0.000000	0.000000
max	220848.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	91234.000000	91234.0	91234.0	91234.0	91234.0	91234.0	91234.0
mean	112575.311989	0.0	0.0	0.0	0.0	0.0	0.0
std	63780.718361	0.0	0.0	0.0	0.0	0.0	0.0
min	0.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	57420.500000	0.0	0.0	0.0	0.0	0.0	0.0
50%	114303.000000	0.0	0.0	0.0	0.0	0.0	0.0
75%	167940.000000	0.0	0.0	0.0	0.0	0.0	0.0
max	220852.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	25095.000000	25095.0	25095.0	25095.0	25095.0	25095.0	25095.0
mean	113323.783702	0.0	0.0	0.0	0.0	0.0	0.0
std	62525.957493	0.0	0.0	0.0	0.0	0.0	0.0
min	1.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	59747.000000	0.0	0.0	0.0	0.0	0.0	0.0
50%	116830.000000	0.0	0.0	0.0	0.0	0.0	0.0
75%	167029.500000	0.0	0.0	0.0	0.0	0.0	0.0
max	220851.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	195758.000000	195758.000000	195758.000000	195758.000000	195758.000000	195758.000000	195758.000000
mean	110054.521532	0.750595	0.163104	0.068922	0.049158	0.017123	0.023197
std	63901.404852	0.432670	0.369462	0.253322	0.216198	0.129731	0.150529
min	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	54702.250000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
50%	109516.500000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
75%	165445.500000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
max	220852.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	10104.000000	10104.0	10104.0	10104.0	10104.0	10104.0	10104.0
mean	115568.118864	0.0	0.0	0.0	0.0	0.0	0.0
std	62281.130722	0.0	0.0	0.0	0.0	0.0	0.0
min	24.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	62664.000000	0.0	0.0	0.0	0.0	0.0	0.0
50%	121571.500000	0.0	0.0	0.0	0.0	0.0	0.0
75%	168562.500000	0.0	0.0	0.0	0.0	0.0	0.0
max	220849.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	6771.000000	6771.0	6771.0	6771.0	6771.0	6771.0	6771.0
mean	114612.731354	0.0	0.0	0.0	0.0	0.0	0.0
std	62597.833021	0.0	0.0	0.0	0.0	0.0	0.0
min	24.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	61470.500000	0.0	0.0	0.0	0.0	0.0	0.0
50%	120668.000000	0.0	0.0	0.0	0.0	0.0	0.0
75%	168218.500000	0.0	0.0	0.0	0.0	0.0	0.0
max	220849.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	5475.000000	5475.0	5475.0	5475.0	5475.0	5475.0	5475.0
mean	113767.193973	0.0	0.0	0.0	0.0	0.0	0.0
std	62271.146454	0.0	0.0	0.0	0.0	0.0	0.0
min	24.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	61459.500000	0.0	0.0	0.0	0.0	0.0	0.0
50%	119135.000000	0.0	0.0	0.0	0.0	0.0	0.0
75%	166385.000000	0.0	0.0	0.0	0.0	0.0	0.0
max	220849.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	4225.000000	4225.0	4225.0	4225.0	4225.0	4225.0	4225.0
mean	113050.889941	0.0	0.0	0.0	0.0	0.0	0.0
std	61702.225308	0.0	0.0	0.0	0.0	0.0	0.0
min	24.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	61077.000000	0.0	0.0	0.0	0.0	0.0	0.0
50%	118479.000000	0.0	0.0	0.0	0.0	0.0	0.0
75%	163889.000000	0.0	0.0	0.0	0.0	0.0	0.0
max	220765.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	219233.000000	219233.000000	219233.000000	219233.000000	219233.000000	219233.000000	219233.000000
mean	110384.700013	0.743907	0.150073	0.069337	0.050230	0.016667	0.023664
std	63759.941782	0.436475	0.357144	0.254027	0.218419	0.128021	0.152001
min	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	55170.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
50%	110307.000000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
75%	165615.000000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
max	220852.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	2796.000000	2796.0	2796.0	2796.0	2796.0	2796.0	2796.0
mean	113590.826538	0.0	0.0	0.0	0.0	0.0	0.0
std	62177.845272	0.0	0.0	0.0	0.0	0.0	0.0
min	84.000000	0.0	0.0	0.0	0.0	0.0	0.0
25%	61212.750000	0.0	0.0	0.0	0.0	0.0	0.0
50%	117528.500000	0.0	0.0	0.0	0.0	0.0	0.0
75%	166869.500000	0.0	0.0	0.0	0.0	0.0	0.0
max	220775.000000	0.0	0.0	0.0	0.0	0.0	0.0

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	373513.000000	373513.000000	373513.000000	373513.000000	373513.000000	373513.000000	373513.000000
mean	109234.044371	0.389732	0.053401	0.028318	0.021341	0.005207	0.010383
std	63928.644515	0.487690	0.224832	0.165879	0.144517	0.071974	0.101364
min	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	53869.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
50%	107649.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
75%	164715.000000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
max	220852.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000

	Unnamed: 0	Growth	Inflation	Cad	Money	Fiscal	Other
count	179788.000000	179788.000000	179788.000000	179788.000000	179788.000000	179788.000000	179788.000000
mean	109290.296388	0.836074	0.100212	0.086813	0.049469	0.004555	0.017409
std	64015.937259	0.370210	0.300284	0.281562	0.216847	0.067340	0.130791
min	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	53842.750000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
50%	107597.500000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
75%	164891.250000	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000
max	220852.000000	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000