def Get_Earnings_Surprise(stock):
'''Go to Yahoo Finance and scrape the Analyst Estimates. Determine if the Earning Suprises were positive.
Return the results in a JSON object.'''
analyst_estimates = 'http://ca.finance.yahoo.com/q/ae?s=' + stock + '+Analyst+Estimates'
#Get Last 4 EPS Surprises
analyst_estimates_web = requests.get(analyst_estimates)
analyst_estimates_text = analyst_estimates_web.content
analyst_estimates_soup = bs(analyst_estimates_text, "lxml")
analyst_estimates_rows = analyst_estimates_soup.findAll('tr')
#i = 0
try:
for n in analyst_estimates_rows:
n = str(n)
if mods.Match('<tr><td class="yfnc_tablehead1"(.*?)EPS Est', n):
n = n.split('<td class="yfnc_tabledata1">')
est_list = [n[1], n[2], n[3], n[4]]
for lst, item in enumerate(est_list):
est_list[lst] = float(item.replace('</tr>', '').replace('</td>', ''))
elif mods.Match('<tr><td class="yfnc_tablehead1"(.*?)EPS Actual', n):
n = n.split('<td class="yfnc_tabledata1">')
actual_list = [n[1], n[2], n[3], n[4]]
for lst, item in enumerate(actual_list):
actual_list[lst] = float(item.replace('</tr>', '').replace('</td>', ''))
elif mods.Match('<tr><th class="yfnc_tablehead1"(.*?)Revenue Est', n):
n = n.split('<th class="yfnc_tablehead1" scope="col" style="font-size:11px;text-align:right;" width="18%">')
dates = [n[1], n[2], n[3], n[4]]
for lst, item in enumerate(dates):
dates[lst] = item.replace('</tr>', '').replace('</td>', '').replace('</th>', '').replace('<br/>', ' ')
surprise1, surprise2, surprise3, surprise4 = 1 - est_list[0]/actual_list[0], 1 - est_list[1]/actual_list[1], 1 - est_list[2]/actual_list[2], 1 - est_list[3]/actual_list[3]
#Test if Earnings Surprise was Positive
is_postive_pes_surprise = False
if surprise1 > 0 and surprise2 > 0 and surprise3 > 0 and surprise4 > 0:
is_postive_pes_surprise = True
#Create Ditionary of Results
results = collections.OrderedDict()
results['Stock'] = stock
results['DateRun'] = mods.Today()
results['TestRun'] = 'Earnings_Surprise'
results[dates[0].replace('Qtr.', '').replace('Current', '',).replace('Next', '').replace('Year', '').strip()] = surprise1
results[dates[1].replace('Qtr.', '').replace('Current', '',).replace('Next', '').replace('Year', '').strip()] = surprise2
results[dates[2].replace('Qtr.', '').replace('Current', '',).replace('Next', '').replace('Year', '').strip()] = surprise3
results[dates[3].replace('Qtr.', '').replace('Current', '',).replace('Next', '').replace('Year', '').strip()] = surprise4
results['Actuals'] = actual_list
results['EST'] = est_list
results['Result'] = is_postive_pes_surprise
return results
except:
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'Earnings_Surprise'
results['DateRun'] = mods.Today()
results['Result'] = 'Test Failed'
return results
def Get_Revenue(stock):
'''Go to Yahoo Finance and Pull the Last Three Years of Revenue Results and Determine if Revenue Passes or Fails Test.
Return the results as a JSON object.'''
revenue = 'http://ca.finance.yahoo.com/q/is?s=' + stock + '&annual'
#Get Revenue Data from Last Income Statement
revenue_web = requests.get(revenue)
revenue_text = revenue_web.content
revenue_soup = bs(revenue_text, "lxml")
try:
# Get Last Three Years of Revenue
revenue = []
for n in revenue_soup.find_all('tr'):
n = str(n)
if mods.Match('<tr><td colspan="2">\n<strong>\n\s+Total Revenue', n):
n = n.replace('\n', '').replace('</strong>', '').replace('</td>', '').replace('<td align="right">', '').replace('</tr', '').strip()
n = n.split('<strong>')
for item in n:
item = item.strip()
if mods.Match('\d+', item):
item = item.replace(' ', '').replace('>', '').replace(',', '').replace('\xc2\xa0\xc2\xa0', '').strip()
revenue.append(item)
last_reported_rev, second_reported_rev, third_reported_rev = float(revenue[0]), float(revenue[1]), float(revenue[2])
#
# #Get Dates for Last Three Reported Years
dates = mods.get_reported_dates(stock)
last_reported_dt, year_before_reported_dt, two_yr_before_reported_dt = dates[0], dates[1], dates[2]
# #Test if Revenue Results are Positive
is_positive_rev = False
if last_reported_rev > second_reported_rev and second_reported_rev > third_reported_rev:
is_positive_rev = True
#
# #Create Ditionary of Results
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'Revenue_Test'
results['DateRun'] = mods.Today()
results['Result'] = is_positive_rev
results[last_reported_dt] = last_reported_rev
results[year_before_reported_dt] = second_reported_rev
results[two_yr_before_reported_dt] = third_reported_rev
return results
except:
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'Revenue_Test'
results['DateRun'] = mods.Today()
results['Result'] = 'Test Failed'
return results
def Get_Earnings_V_Industry(stock):
'''Go to Yahoo finance and pull information from Analyst Estimates. Determine if the earnings
are favorable compared to the industry. Return the results as a JSON object.'''
analyst_estimates = 'http://ca.finance.yahoo.com/q/ae?s=' + stock + '+Analyst+Estimates'
#Get Last 4 EPS Surprises
analyst_estimates_web = requests.get(analyst_estimates)
analyst_estimates_text = analyst_estimates_web.content
analyst_estimates_soup = bs(analyst_estimates_text)
analyst_estimates_rows = analyst_estimates_soup.findAll('tr')
#i = 0
try:
for n in analyst_estimates_rows:
n = str(n)
if mods.Match('<tr><td class="yfnc_tablehead1"(.*?)Price/Earnings',
n):
n = n.split('<td class="yfnc_tabledata1">')
pe_list = [n[1], n[2], n[3], n[4]]
for lst, item in enumerate(pe_list):
pe_list[lst] = float(
item.replace('</tr>', '').replace('</td>',
'').replace('%', ''))
#Test if Earnings Surprise was Positive
is_postive_pe_v_industry = False
if pe_list[0] > pe_list[1]:
is_postive_pe_v_industry = True
#Create Ditionary of Results
results = collections.OrderedDict()
results['Stock'] = stock
results['DateRun'] = mods.Today()
results['TestRun'] = 'Earnings Versus Industry'
results['PE Estimate: ' + stock] = pe_list[0]
results['Industry Estimate:'] = pe_list[1]
results['Result'] = is_postive_pe_v_industry
return results
except:
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'Earnings Versus Industry'
results['DateRun'] = mods.Today()
results['Result'] = 'Test Failed'
return results
def Get_Growth_Forecast(stock):
'''Go to Yahoo Finance and get the analyst estimates. Determine if the last 4 EPS surprise results
were positive. Return the results as a JSON object.'''
analyst_estimates = 'http://ca.finance.yahoo.com/q/ae?s=' + stock + '+Analyst+Estimates'
#Get Last 4 EPS Surprises
analyst_estimates_web = requests.get(analyst_estimates)
analyst_estimates_text = analyst_estimates_web.content
analyst_estimates_soup = bs(analyst_estimates_text, 'lxml')
analyst_estimates_rows = analyst_estimates_soup.findAll('tr')
#i = 0
try:
for n in analyst_estimates_rows:
n = str(n)
if mods.Match('<tr><td class="yfnc_tablehead1"(.*?)Next 5 Years',
n):
n = n.split('<td class="yfnc_tabledata1">')
est_list = [n[1], n[2], n[3], n[4]]
for lst, item in enumerate(est_list):
est_list[lst] = float(
item.replace('</tr>', '').replace('</td>',
'').replace('%', ''))
#Test if Earnings Surprise was Positive
is_postive_growth_forecast = False
if est_list[0] > 8:
is_postive_growth_forecast = True
#Create Ditionary of Results
results = collections.OrderedDict()
results['Stock'] = stock
results['DateRun'] = mods.Today()
results['TestRun'] = 'Growth_Forecast'
results['Growth Forecast: ' + stock] = est_list[0]
results['Result'] = is_postive_growth_forecast
return results
except:
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'Growth_Forecast'
results['DateRun'] = mods.Today()
results['Result'] = 'Test Failed'
return results
def Get_Shares_Short(stock):
'''Got to Yahoo Finance and determine the Insider Tracking of the stock. Return the results as a
JSON object.'''
insider_transactions = 'http://ca.finance.yahoo.com/q/it?s=' + stock + '+Insider+Transactions'
#Get Insider Transaction Data (Last 6 Months and Last 3 Months)
insider_transactions_web = requests.get(insider_transactions)
insider_transactions_text = insider_transactions_web.content
insider_transactions_soup = bs(insider_transactions_text)
insider_transactions_rows = insider_transactions_soup.findAll('tr')
i = 0
insider_list = []
for n in insider_transactions_rows:
if n.text[0:10] == 'Net Shares':
n = str(n)
n = n.split('<td align="right" class="yfnc_tabledata1"')
insider_list.append(n[1].replace('</td>', '').replace(
'</tr>', '').replace(' width="20%">',
'').replace('>', '').replace(',', ''))
try:
#Test if Positive Insider Tracking (Positive Purchase of Stock in Last 3 Months)
is_postive_insider = False
if mods.Match('\(\d+\)', insider_list[1]):
is_postive_insider = True
#Create Ditionary of Results
results = collections.OrderedDict()
results['Stock'] = stock
results['DateRun'] = mods.Today()
results['TestRun'] = 'Insider_Trading'
results['Insider Purchasing 6 Month: ' + stock] = insider_list[0]
results['Insider Purchasing 3 Month: ' + stock] = insider_list[1]
results['Result'] = is_postive_insider
return results
except:
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'Insider_Trading'
results['DateRun'] = mods.Today()
results['Result'] = 'Test Failed'
return results
def Get_Earnings_Forecast(stock):
'''Go to Yahoo Finance and scrape the Analyst Estimates. Determine if the Earnings Forecast is postive.
Return the results in a JSON object.'''
forecast = 'http://ca.finance.yahoo.com/q/ae?s=' + stock + '+Analyst+Estimates'
#Get Last 4 Quarters of Earnings Forecasts
forecast_web = requests.get(forecast)
forecast_text = forecast_web.content
forecast_soup = bs(forecast_text)
forecast_rows = forecast_soup.findAll('tr')
#i = 0
try:
lists = []
for n in forecast_rows:
n = str(n)
if mods.Match('<tr><th class="yfnc_tablehead1"(.*?)Earnings Est',
n):
n = n.split(
'<th class="yfnc_tablehead1" scope="col" style="font-size:11px; font-weight:normal;text-align:right;" width="18%">'
)
dates_list = [n[1], n[2], n[3], n[4]]
for lst, item in enumerate(dates_list):
dates_list[lst] = item.replace('</tr>', '').replace(
'</td>', '').replace('</th>',
'').replace('<br/>', ' ')
elif mods.Match(
'<tr><td class="yfnc_tablehead1"(.*?)Avg. Estimate', n):
n = n.split('<td class="yfnc_tabledata1">')
forecast_list = [n[1], n[2], n[3], n[4]]
try:
for lst, item in enumerate(forecast_list):
forecast_list[lst] = float(
item.replace('</tr>', '').replace('</td>', ''))
lists.append(forecast_list)
except:
dummy = 0
#Test for Positive Forecasts
is_positive_earnings_forecast = False
if lists[0][0] < lists[0][1] and lists[0][1] < lists[0][2] and lists[
0][2] < lists[0][3]:
is_positive_earnings_forecast = True
#Create Ditionary of Results
results = collections.OrderedDict()
results['Stock'] = stock
results['TestRun'] = 'Earnings_Forecast'
results['DateRun'] = mods.Today()
results['Result'] = is_positive_earnings_forecast
results['Current Qtr'] = lists[0][0]
results['Next Qtr'] = lists[0][1]
results['Current Year'] = lists[0][2]
results['Next Year'] = lists[0][3]
return results
except:
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'Earnings_Forecast'
results['DateRun'] = mods.Today()
results['Result'] = 'Test Failed'
return results
##Testing:
#stock = 'AAPL'
#print Get_Earnings_Forecast(stock)
def Get_ROE(stock):
'''Go to Yahoo Finance. Get the results from the Key Statistics page, Income Statement, and Balance Sheet.
Determine if the results are positive for a good return on equity. Return the results as a JSON object.'''
key_stats = 'http://ca.finance.yahoo.com/q/ks?s=' + stock + '+Key+Statistics'
income_statement = 'http://ca.finance.yahoo.com/q/is?s=' + stock + '+Income+Statement&annual'
balance_sheet = 'http://ca.finance.yahoo.com/q/bs?s=' + stock + '+Balance+Sheet&annual'
#Get Last 12 Months ROE
key_stats_web = requests.get(key_stats)
key_stats_text = key_stats_web.content
key_stats_soup = bs(key_stats_text, "lxml")
for n in key_stats_soup.find_all('tr'):
if n.text[0:16] == 'Return on Equity':
amount = n.text.split(':')
amount = str(amount[1])
roe_12month = float(amount[0:-1])
# Get Last Three Years of Return on Equity (Net Income/Stock Holder's Equity)
# First get the Last Three Years of Net Income from Income Statement
income_statement_web = requests.get(income_statement)
income_statement_text = income_statement_web.content
income_statement_soup = bs(income_statement_text)
income_statement_rows = income_statement_soup.findAll('tr')
#i = 0
try:
for row in income_statement_rows:
row = str(row)
if mods.Match('<tr><td colspan="2">\n<strong>\n\s+Net Income', row):
row = row.split('\n')
last_reported_ni, second_reported_ni, third_reported_ni = row[6].strip(), row[10].strip(), row[14].strip()
last_reported_ni, second_reported_ni, third_reported_ni = float(last_reported_ni[0:-4].replace(',', '')),float(second_reported_ni[0:-4].replace(',', '')), float(third_reported_ni[0:-4].replace(',', ''))
# Second get the Last Three Years of StockHolder's Equity from Balance Sheet
balance_sheet_web = requests.get(balance_sheet)
balance_sheet_text = balance_sheet_web.content
balance_sheet_soup = bs(balance_sheet_text, 'lxml')
balance_sheet_rows = balance_sheet_soup.findAll('tr')
for row in balance_sheet_rows:
row = str(row)
if mods.Match('<tr><td colspan="2">\n<strong>\n\s+Total Stockholder Equity', row):
row = row.split('\n')
last_reported_se, second_reported_se, third_reported_se = row[6].strip(), row[10].strip(), row[14].strip()
last_reported_se, second_reported_se, third_reported_se = float(last_reported_se[0:-4].replace(',', '')),float(second_reported_se[0:-4].replace(',', '')), float(third_reported_se[0:-4].replace(',', ''))
#Third Calculate ROE (Net Income/Stock Holders Equity)
last_reported_roe, second_reported_roe, third_reported_roe = last_reported_ni/last_reported_se, second_reported_ni/second_reported_se, third_reported_ni/third_reported_se
#Get Dates for Last Three Reported Years
last_reported_dt, year_before_reported_dt, two_yr_before_reported_dt = mods.get_reported_dates(stock)
is_positive_rev = False
now = datetime.datetime.now()
if int(last_reported_dt[0:2]) - now.month < 6:
if roe_12month > second_reported_roe and second_reported_roe > third_reported_roe:
is_positive_rev = True
# Time since last annual reported revenue was less than 6 months
else:
if last_reported_roe > second_reported_roe and second_reported_roe > third_reported_roe:
is_positive_rev = True
#Create Ditionary of Results
results = collections.OrderedDict()
results['Stock'] = stock
results['DateRun'] = mods.Today()
results['TestRun'] = 'Positive_Return_on_Equity'
results['Result'] = is_positive_rev
results[last_reported_dt] = last_reported_roe
results[year_before_reported_dt] = second_reported_roe
results[two_yr_before_reported_dt] = third_reported_roe
results['12 Month Return On Equity'] = roe_12month
return results
except:
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'Positive_Return_on_Equity'
results['DateRun'] = mods.Today()
results['Result'] = 'Test Failed'
return results
def Get_EPS(stock):
#Go to Yahoo Finance and Pull the Last Three Years of Revenue Results and Determine if Revenue Passes or Fails Test
#Current Year: http://finance.yahoo.com/q/ks?s=AAPL+Key+Statistics
#Last Years Income Statement: http://finance.yahoo.com/q/is?s=AAPL+Income+Statement&annual
key_stats = 'http://ca.finance.yahoo.com/q/ks?s=' + stock + '+Key+Statistics'
market_watch = 'http://www.marketwatch.com/investing/stock/' + stock + '/financials'
#Get Last 12 Months Earnings Per Share
key_stats_web = requests.get(key_stats)
key_stats_text = key_stats_web.content
key_stats_soup = bs(key_stats_text)
for n in key_stats_soup.find_all('tr'):
if n.text[0:11] == 'Diluted EPS':
amount = n.text.split(':')
eps_12month = float(amount[1])
# Get Last Three Years of Earnings Per Share
market_watch_web = requests.get(market_watch)
market_watch_text = market_watch_web.content
market_watch_soup = bs(market_watch_text)
market_watch_rows = market_watch_soup.findAll('tr')
#tag = '<tr class="mainRow"> \
#<td class="rowTitle"><a data-ref="ratio_Eps1YrAnnualGrowth" href="#"><span class="expand"></span></a> EPS (Basic)</td>'
#i = 0
try:
for row in market_watch_rows:
row = str(row)
if mods.Match('<tr class="mainRow">\n(.*?)EPS \(Diluted\)', row):
row = row.split('<td class="valueCell">')
last_reported_eps, second_reported_eps, third_reported_eps = float(
row[5][0:4].strip()), float(row[4][0:4].strip()), float(
row[3][0:4].strip())
#Get Dates for Last Three Reported Years
last_reported_dt, year_before_reported_dt, two_yr_before_reported_dt = mods.get_reported_dates(
stock)
#print mods.get_reported_dates(stock)
is_positive_eps = False
now = datetime.datetime.now()
if int(last_reported_dt[0:2]) - now.month < 6:
if eps_12month > second_reported_eps and second_reported_eps > third_reported_eps:
is_positive_eps = True
# Time since last annual reported revenue was less than 6 months
else:
if last_reported_eps > second_reported_eps and second_reported_eps > third_reported_eps:
is_positive_eps = True
'''
'''
#Create Ditionary of Results
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'EPS'
results['DateRun'] = mods.Today()
results['Result'] = is_positive_eps
results[last_reported_dt] = last_reported_eps
results[year_before_reported_dt] = second_reported_eps
results[two_yr_before_reported_dt] = third_reported_eps
results['12 Month Reported EPS'] = eps_12month
return results
except:
results = collections.OrderedDict()
results['stock'] = stock
results['TestRun'] = 'EPS'
results['DateRun'] = mods.Today()
results['Result'] = 'Test Failed'
return results
#Testing:
#stock = 'AAPL'
#print Get_EPS(stock)
