ANOVA for Feature Selection

# !pip install dataidea==0.2.5

import scipy as sp
from sklearn.feature_selection import SelectKBest
from dataidea.datasets import loadDataset

fpl = loadDataset('fpl') # load fpl inbuilt

fpl.head(n=5) # select top 5

	First_Name	Second_Name	Club	Goals_Scored	Assists	Total_Points	Minutes	Saves	Goals_Conceded	Creativity	Influence	Threat	Bonus	BPS	ICT_Index	Clean_Sheets	Red_Cards	Yellow_Cards	Position
0	Bruno	Fernandes	MUN	18	14	244	3101	0	36	1414.9	1292.6	1253	36	870	396.2	13	0	6	MID
1	Harry	Kane	TOT	23	14	242	3083	0	39	659.1	1318.2	1585	40	880	355.9	12	0	1	FWD
2	Mohamed	Salah	LIV	22	6	231	3077	0	41	825.7	1056.0	1980	21	657	385.8	11	0	0	MID
3	Heung-Min	Son	TOT	17	11	228	3119	0	36	1049.9	1052.2	1046	26	777	315.2	13	0	0	MID
4	Patrick	Bamford	LEE	17	11	194	3052	0	50	371.0	867.2	1512	26	631	274.6	10	0	3	FWD

# Create groups of goals scored for each player position

forwards_goals = fpl[fpl.Position == 'FWD']['Goals_Scored']
midfielders_goals = fpl[fpl.Position == 'MID']['Goals_Scored']
defenders_goals = fpl[fpl.Position == 'DEF']['Goals_Scored']
goalkeepers_goals = fpl[fpl.Position == 'GK']['Goals_Scored']

# Perform the ANOVA test for the groups

f_statistic, p_value = sp.stats.f_oneway(forwards_goals, midfielders_goals,
                                         defenders_goals, goalkeepers_goals
                                        )
print("F-statistic:", f_statistic)
print("p-value:", p_value)

F-statistic: 33.281034594400445
p-value: 3.9257634156019246e-20

# Create groups of assists for each player position

forwards_assists = fpl[fpl.Position == 'FWD']['Assists']
midfielders_assists = fpl[fpl.Position == 'MID']['Assists']
defenders_assists = fpl[fpl.Position == 'DEF']['Assists']
goalkeepers_assists = fpl[fpl.Position == 'GK']['Assists']

# Perform the ANOVA test for the groups

f_statistic, p_value = sp.stats.f_oneway(forwards_assists, midfielders_assists,
                                         defenders_assists, goalkeepers_assists
                                        )
print("F-statistic:", f_statistic)
print("p-value:", p_value)

F-statistic: 19.263717036430815
p-value: 5.124889288362087e-12

# Use scikit-learn's SelectKBest (with f_classif)

test = SelectKBest(k=1)
fit = test.fit(fpl[['Goals_Scored', 'Assists']], fpl.Position)
scores = fit.scores_
features = fit.transform(fpl[['Goals_Scored', 'Assists']])
selected_indices = test.get_support(indices=True)

print('Feature Scores: ', scores)
print('Selected Features Indices: ', selected_indices)

Feature Scores:  [33.28103459 19.26371704]
Selected Features Indices:  [0]