def advisor1(u_in, data, economy, show=True):
score = {}
criterions = []
criterions.append('Security')
desc_df = pd.DataFrame()
### score anr
if (1 in u_in):
score = ut.score_feature(data, score, 'ANR')
criterions.append('ANR')
if (2 in u_in):
score = ut.score_feature(data, score, 'Dividents')
criterions.append('Dividents')
### score P/E
if (3 in u_in):
score = ut.score_feature(data, score, 'PE')
criterions.append('PE')
### score returns last 3 months
if (4 in u_in):
score = ut.score_feature(data, score, 'Returns_3_months')
criterions.append('Returns_3_months')
if (5 in u_in):
score = ut.score_feature(data, score, 'Ethics')
criterions.append('Ethics')
if (6 in u_in):
score = ut.score_feature(data, score, 'Bribery')
criterions.append('Bribery')
### score sectors from BC
if (7 in u_in):
sectors = ut.check_bc(economy)
score = ut.score_sectors(data, sectors, score)
criterions.append('Sector')
### determine ranking
rank = ut.rank_scores(score)
### make df
reasoning = pd.DataFrame(list(rank.items()), columns=['Company', 'Score'])
reasoning = reasoning.set_index('Company')
desc_df = data[criterions]
desc_df = desc_df.set_index('Security')
ranking = pd.concat([reasoning, desc_df], axis=1)
ranking = ranking.sort_values(by='Score', axis=0, ascending=False)
criterions.remove('Security')
if show:
print(ranking.head(25))
return ranking
def rank_by_preferences(u_in, data, economy, show=True):
score = {}
criterions = []
criterions.append('Name')
desc_df = pd.DataFrame()
### score anr
if (1 in u_in):
score = ut.score_feature(data, score, 'Working_Capital')
criterions.append('Working_Capital')
if (2 in u_in):
score = ut.score_feature(data, score, 'Fixed_Assets_Turnover')
criterions.append('Fixed_Assets_Turnover')
### score P/E
if (3 in u_in):
score = ut.score_feature(data, score, 'Accounts_Receivable')
criterions.append('Accounts_Receivable')
### score returns last 3 months
if (4 in u_in):
score = ut.score_feature(data, score, 'Revenue')
criterions.append('Revenue')
if (5 in u_in):
score = ut.score_feature(data, score, 'revenue_3Y_Avg')
criterions.append('revenue_3Y_Avg')
if (6 in u_in):
score = ut.score_feature(data, score, 'Free_Cash_Flow_Per_Share')
criterions.append('Free_Cash_Flow_Per_Share')
### score sectors from BC
if (7 in u_in):
sectors = ut.check_bc(economy)
score = ut.score_sectors(data, sectors, score)
criterions.append('Sector')
### determine ranking
rank = ut.rank_scores(score)
### make df
reasoning = pd.DataFrame(list(rank.items()), columns=['Name', 'Score'])
reasoning = reasoning.set_index('Name')
desc_df = data[criterions]
#desc_df = desc_df.set_index('Symbol')
ranking = pd.concat([reasoning, desc_df], axis=1)
ranking = ranking.sort_values(by='Score', axis=0, ascending=False)
#criterions.remove('Symbol')
if show:
print(ranking.head(25))
return ranking
def rank_stocks(economy):
data = pd.read_hdf('stocks.hdf5', 'Datataset1/X')
data = data.dropna()
scor = {}
rank = {}
scor = ut.score_feature(data, scor, 'ANR')
scor = ut.score_feature(data, scor, 'Returns_3_months')
scor = ut.score_feature(data, scor, 'PE')
sectors = ut.check_bc(economy)
scor = ut.score_sectors(data, sectors, scor)
rank = ut.rank_scores(scor)
rank = pd.DataFrame(list(rank.items()), columns=['Name', 'Score'])
rank = rank.sort_values(by='Score', axis=0, ascending=False)
securities = rank['Name'].values.tolist()
return securities
score = ut.score_feature(data, score, 'ANR')
criterions.append('ANR')
### score returns last 3 months
if (2 in user_in):
score = ut.score_feature(data, score, 'Returns_3_months')
criterions.append('Returns_3_months')
### score P/E
if (3 in user_in):
score = ut.score_feature(data, score, 'PE')
criterions.append('PE')
### score sectors
if (4 in user_in):
sectors = ut.check_bc(economy)
score = ut.score_sectors(data, sectors, score)
criterions.append('Sector')
### determine ranking
rank = ut.rank_scores(score)
### make df
reasoning = pd.DataFrame(list(rank.items()), columns=['Company', 'Score'])
reasoning = reasoning.set_index('Company')
desc_df = data[criterions]
desc_df = desc_df.set_index('Security')
ranking = pd.concat([reasoning, desc_df], axis=1)
ranking = ranking.sort_values(by='Score', axis=0, ascending=False)