Sustainable Portfolio Selection - An Approach To Impact Modelling

    	
            
import pandas
import numpy
import random
import matplotlib.pyplot as plt
import scipy.optimize as sco 
import os
import refinitiv.dataplatform.eikon as ek
 
ek.set_app_key('YOUR APPKEY HERE')

        
        Copy
    

    	
            
N = 350
constituents, err = ek.get_data(['IWDA.L'], ['TR.ETPConstituentRIC', 'TR.ETPConstituentName'])
constituents.rename(columns={'Constituent RIC': 'ric', 'Constituent Name': 'name'}, inplace=True)
constituents = constituents[['ric','name']][0:N].drop_duplicates(subset=['ric'])
ric_list = list(constituents.ric[constituents.ric.astype(bool)])
 
df_esg, err = ek.get_data(ric_list, ['TR.CommonName', 'TR.TRESGScore','TR.BusinessSummary','TR.NACEClassification'])
df_esg = df_esg.rename(columns={'Company Common Name':'name', 'Instrument':'ric', 'ESG Score':'esg', 'NACE Classification':'nace'})
df_esg = df_esg.drop_duplicates(subset=['ric']).set_index('ric')

        
        Copy
    

    	
            
nace_code_list =[]
for instr in ric_list:
    df_instr = df_esg.loc[instr]
    nace_code_list.append(df_instr['nace'][-6:-4])
    
df_nace = pandas.DataFrame(data=[ric_list,nace_code_list],index=['ric','nace_code']).T.set_index('ric')
df_esg_nace = pandas.concat([df_esg,df_nace],axis=1)[df_nace['nace_code'].map(len)>0]

        
        Copy
    

    	
            
esg_bound = 70
df_esg_constraint = df_esg_nace[(df_esg_nace.esg >= esg_bound)].replace('','nan')
K = len(df_esg_constraint.index)
print('Selected {K} out of {N} instruments by filtering ESG Score >= {esg}\n'.format(K=K,N=N,esg=esg_bound))
df_esg_constraint

        
        Copy
    

    	
            
print('Loading timeseries data from eikon')
start='2020-01-01'
end='2020-12-31'
instruments = df_esg_constraint.index
timeseries_data =pandas.DataFrame()
for r in instruments:
    try:
        ts1 = ek.get_timeseries(r,'CLOSE',start_date=start,end_date=end,interval='daily')
        ts1.rename(columns = {'CLOSE' : r}, inplace=True)
        timeseries_data =pandas.concat([timeseries_data, ts1], axis=1)
    except:
        continue
 
timeseries_data = timeseries_data.dropna()
timeseries_data

        
        Copy
    

    	
            
returns = timeseries_data.pct_change().replace(numpy.inf, numpy.nan).dropna()
covMatrix = returns.cov()
 
def risk_measure(covMatrix, weights):
    return numpy.dot(weights, numpy.dot(covMatrix, weights))
 
bounds = K * [(0, 1)]
 
constraints = {'type': 'eq', 'fun': lambda weights: weights.sum() - 1}
 
mvp = sco.minimize(lambda x: risk_measure(covMatrix, x),  # function to be minimized
                   K * [1 / K],  # initial guess
                   bounds=bounds,  # box constraints
                   constraints =constraints,  # equality constraints
                   )
 
print('''MVP in a universe with {K} instruments
Number of selected instruments: {n}
Minimum weight: {minw}
Maximum weight: {maxw}
Historical risk measure: {risk}
Historical return p.a.: {r}
ESG score: {esg}'''.format(K=K,
                           n=numpy.sum(mvp['x']>1e-4),
                           minw=numpy.min(mvp['x'][numpy.nonzero(mvp['x'])]),
                           maxw=numpy.max(mvp['x']),
                           risk=risk_measure(covMatrix, mvp['x']),
                           r=numpy.dot(mvp['x'],returns.sum()),
                           esg=numpy.dot(mvp['x'], df_esg_constraint['esg'])))

        
        Copy
    

    	
            
nace_dict ={'01': 'Crop and animal production, hunting and related service activities', '02': 'Forestry and logging', '03': 'Fishing and aquaculture', '05': 'Mining of coal and lignite', '06': 'Extraction of crude petroleum and natural gas', '07': 'Mining of metal ores', '08': 'Other mining and quarrying', '09': 'Mining support service activities', '10': 'Manufacture of food products', '11': 'Manufacture of beverages', '12': 'Manufacture of tobacco products', '13': 'Manufacture of textiles', '14': 'Manufacture of wearing apparel', '15': 'Manufacture of leather and related products', '16': 'Manufacture of wood and of products of wood and cork, except furniture; manufacture of articles of straw and plaiting materials', '17': 'Manufacture of paper and paper products', '18': 'Printing and reproduction of recorded media', '19': 'Manufacture of coke and refined petroleum products', '20': 'Manufacture of chemicals and chemical products', '21': 'Manufacture of basic pharmaceutical products and pharmaceutical preparations', '22': 'Manufacture of rubber and plastic products', '23': 'Manufacture of other non-metallic mineral products', '24': 'Manufacture of basic metals', '25': 'Manufacture of fabricated metal products, except machinery and equipment', '26': 'Manufacture of computer, electronic and optical products', '27': 'Manufacture of electrical equipment', '28': 'Manufacture of machinery and equipment n.e.c.', '29': 'Manufacture of motor vehicles, trailers and semi-trailers', '30': 'Manufacture of other transport equipment', '31': 'Manufacture of furniture', '32': 'Other manufacturing', '33': 'Repair and installation of machinery and equipment', '35': 'Electricity, gas, steam and air conditioning supply', '36': 'Water collection, treatment and supply', '37': 'Sewerage', '38': 'Waste collection, treatment and disposal activities; materials recovery', '39': 'Remediation activities and other waste management services', '41': 'Construction of buildings', '42': 'Civil engineering', '43': 'Specialised construction activities', '45': 'Wholesale and retail trade and repair of motor vehicles and motorcycles', '46': 'Wholesale trade, except of motor vehicles and motorcycles', '47': 'Retail trade, except of motor vehicles and motorcycles', '49': 'Land transport and transport via pipelines', '50': 'Water transport', '51': 'Air transport', '52': 'Warehousing and support activities for transportation', '53': 'Postal and courier activities', '55': 'Accommodation', '56': 'Food and beverage service activities', '58': 'Publishing activities', '59': 'Motion picture, video and television programme production, sound recording and music publishing activities', '60': 'Programming and broadcasting activities', '61': 'Telecommunications', '62': 'Computer programming, consultancy and related activities', '63': 'Information service activities', '64': 'Financial service activities, except insurance and pension funding', '65': 'Insurance, reinsurance and pension funding, except compulsory social security', '66': 'Activities auxiliary to financial services and insurance activities', '68': 'Real estate activities', '69': 'Legal and accounting activities', '70': 'Activities of head offices; management consultancy activities', '71': 'Architectural and engineering activities; technical testing and analysis', '72': 'Scientific research and development ', '73': 'Advertising and market research', '74': 'Other professional, scientific and technical activities', '75': 'Veterinary activities', '77': 'Rental and leasing activities', '78': 'Employment activities', '79': 'Travel agency, tour operator and other reservation service and related activities', '80': 'Security and investigation activities', '81': 'Services to buildings and landscape activities', '82': 'Office administrative, office support and other business support activities', '84': 'Public administration and defence; compulsory social security', '85': 'Education', '86': 'Human health activities', '87': 'Residential care activities', '88': 'Social work activities without accommodation', '90': 'Creative, arts and entertainment activities', '91': 'Libraries, archives, museums and other cultural activities', '92': 'Gambling and betting activities', '93': 'Sports activities and amusement and recreation activities', '94': 'Activities of membership organisations', '95': 'Repair of computers and personal and household goods', '96': 'Other personal service activities', '97': 'Activities of households as employers of domestic personnel', '98': 'Undifferentiated goods- and services-producing activities of private households for own use', '99': 'Activities of extraterritorial organisations and bodies', 'nan': 'Others'}
 
def calcWeightsForPieChart(df):
    #calculate weights for pie chart
    df_nan = df.replace('','nan')
    sections = {item for item in list(df_nan['nace_code']) if len(item)>0}
    piesizes = {}
    for s in sections:
        piesizes[str(s)] = df_nan[df_nan['nace_code']==s]['weight'].values.sum()
 
    return piesizes
 
def plotPieChartWithNACELabels(nace_dict,piesizes):
    ## piechart for nace codes
    labels = []
    sizes = []
    for x, y in piesizes.items():
        labels.append('{d} ({c})'.format(d=nace_dict[x],c=str(x)))
        sizes.append(y)
    plt.figure(figsize=(10, 10), dpi=80)
    plt.pie(sizes, labels=labels)
    plt.axis('equal')
    plt.show()
    
df_esg_constraint['weight'] = list(mvp['x'])
piesizes = calcWeightsForPieChart(df_esg_constraint)
plotPieChartWithNACELabels(nace_dict, piesizes)

        
        Copy
    

    	
            
esg_ret_mvp = sco.minimize(lambda x: risk_measure(covMatrix, x) - 0.0002*numpy.dot(x, returns.sum()) - 0.00002*numpy.dot(x, df_esg_constraint['esg']),  # function to be minimized
                   K * [1 / K],  # initial guess
                   bounds=bounds,  # box constraints
                   constraints =constraints,  # equality constraints
                   )
 
print('''Solution to balanced volatility, return, ESG problem in a universe with {K} instruments
Number of selected instruments: {n}
Minimum weight: {minw}
Maximum weight: {maxw}
Historical risk measure: {risk}
Historical return p.a.: {r}
ESG score: {esg}'''.format(K=K,
                           n=numpy.sum(esg_ret_mvp['x']>1e-4),
                           minw=numpy.min(esg_ret_mvp['x'][numpy.nonzero(esg_ret_mvp['x'])]),
                           maxw=numpy.max(esg_ret_mvp['x']),
                           risk=risk_measure(covMatrix, esg_ret_mvp['x']),
                           r=numpy.dot(esg_ret_mvp['x'],returns.sum()),
                           esg=numpy.dot(esg_ret_mvp['x'], df_esg_constraint['esg'])))

        
        Copy
    

    	
            
df_esg_constraint['weight'] = list(esg_ret_mvp['x'])
 
piesizes = calcWeightsForPieChart(df_esg_constraint)
plotPieChartWithNACELabels(nace_dict, piesizes)

        
        Copy
    

    	
            
nace_impact_dict ={key: 0 for key in nace_dict.keys()}
nace_impact_dict['41']=10
#nace_impact_dict ={key: random.randint(0, 10) for key in nace_dict.keys()}
df_esg_constraint['nace_impact'] = df_esg_constraint['nace_code'].apply(lambda x: nace_impact_dict[x])

        
        Copy
    

    	
            impact_esg_score = lambda df: df['nace_impact']*(df['esg']-esg_bound)
        
        Copy
    

    	
            
mvp_impact_esg = numpy.dot(mvp['x'], impact_esg_score(df_esg_constraint))
mvp_ret_impact_esg = numpy.dot(esg_ret_mvp['x'], impact_esg_score(df_esg_constraint))
print('ESG of MVP: {mvpesg}\nImpact-ESG of MVP: {mvpiesg}\n'.format(mvpesg=numpy.dot(mvp['x'], df_esg_constraint['esg']), mvpiesg=mvp_impact_esg))
print('ESG of balanced: {balesg}\nImpact-ESG of balanced: {baliesg}'.format(balesg=numpy.dot(esg_ret_mvp['x'], df_esg_constraint['esg']), baliesg=mvp_ret_impact_esg))

        
        Copy
    

    	
            
impact_esg_mvp = sco.minimize(lambda x: risk_measure(covMatrix, x) - 0.0002*numpy.dot(x, returns.sum()) - 0.000004*numpy.dot(x, impact_esg_score(df_esg_constraint)),  # function to be minimized
                   K * [1 / K],  # initial guess
                   bounds=bounds,  # boundary conditions
                   constraints =constraints,  # equality constraints
                   )
 
 
print('''Solution to weighted impact ESG problem in a universe with {K} instruments
Number of selected instruments: {n}
Minimum weight: {minw}
Maximum weight: {maxw}
Historical risk measure: {risk}
Historical return p.a.: {r}
ESG score: {esg}'''.format(K=K,
                           n=numpy.sum(impact_esg_mvp['x']>1e-4),
                           minw=numpy.min(impact_esg_mvp['x'][numpy.nonzero(impact_esg_mvp['x'])]),
                           maxw=numpy.max(impact_esg_mvp['x']),
                           risk=risk_measure(covMatrix, impact_esg_mvp['x']),
                           r=numpy.dot(impact_esg_mvp['x'],returns.sum()),
                           esg=numpy.dot(impact_esg_mvp['x'], df_esg_constraint['esg'])))

        
        Copy
    

    	
            
mvp_impact_esg = numpy.dot(mvp['x'], impact_esg_score(df_esg_constraint))
mvp_ret_impact_esg = numpy.dot(esg_ret_mvp['x'], impact_esg_score(df_esg_constraint))
 
print('ESG of MVP: {esg}\nImpact-ESG of MVP: {iesg}\n'.format(
    esg=numpy.dot(mvp['x'], df_esg_constraint['esg']), 
    iesg=mvp_impact_esg))
 
print('ESG of return and ESG balanced: {esg}\nImpact-ESG of return and ESG balanced: {iesg}\n'.format(
    esg=numpy.dot(esg_ret_mvp['x'], df_esg_constraint['esg']),
    iesg=mvp_ret_impact_esg))
print('ESG of return and impact-ESG balanced: {esg}\nImpact-ESG of return and impact-ESG balanced: {iesg}\n'.format(
    esg=numpy.dot(impact_esg_mvp['x'], df_esg_constraint['esg']), 
    iesg=numpy.dot(impact_esg_mvp['x'], impact_esg_score(df_esg_constraint))))

        
        Copy
    

    	
            
df_esg_constraint['weight'] = impact_esg_mvp['x']
 
piesizes = calcWeightsForPieChart(df_esg_constraint)
plotPieChartWithNACELabels(nace_dict, piesizes)

        
        Copy