def update_historical_fly_time_series_market_commodity_futures_tab3(
rows_data, rows_selected, lookback_window):
sym_root = rows_data[rows_selected[0]]['Name']
leg1 = rows_data[rows_selected[0]]['Leg1 Actual']
leg2 = rows_data[rows_selected[0]]['Leg2 Actual']
leg3 = rows_data[rows_selected[0]]['Leg3 Actual']
lookback_date = convert_date_input(lookback_window, datetime(2008, 1, 1))
if ':' in sym_root:
df1 = inter_comdty_spread_hist_data_dict[sym_root][leg1]
df2 = inter_comdty_spread_hist_data_dict[sym_root][leg2]
df3 = inter_comdty_spread_hist_data_dict[sym_root][leg3]
else:
df1 = futures_hist_prices_dict[sym_root][leg1]
df2 = futures_hist_prices_dict[sym_root][leg2]
df3 = futures_hist_prices_dict[sym_root][leg3]
df = df1 - df2 * 2 + df3
df = df[lookback_date.date():]
trace = go.Scatter(x=df.index, y=df, name=f'{leg1}-{leg2}', mode='lines')
layout_fig = go.Layout(title=sym_root,
xaxis={
'title': sym_root,
'type': 'date',
'tickformat': '%Y-%m-%d'
},
yaxis={'title': 'Price'},
legend=dict(orientation="h"),
paper_bgcolor='rgba(0,0,0,0)',
plot_bgcolor='rgba(0,0,0,0)')
return go.Figure(data=[trace], layout=layout_fig)
def update_historical_fly_scatterplot_market_commodity_futures_tab3(
rows_data, rows_selected, lookback_window):
sym_root = rows_data[rows_selected[0]]['Name']
leg1 = rows_data[rows_selected[0]]['Leg1 Actual']
leg2 = rows_data[rows_selected[0]]['Leg2 Actual']
leg3 = rows_data[rows_selected[0]]['Leg3 Actual']
lookback_date = convert_date_input(lookback_window, datetime(2008, 1, 1))
if ':' in sym_root:
df1 = inter_comdty_spread_hist_data_dict[sym_root][leg1]
df2 = inter_comdty_spread_hist_data_dict[sym_root][leg2]
df3 = inter_comdty_spread_hist_data_dict[sym_root][leg3]
df0 = generic_inter_comdty_hist_prices_dict[sym_root][sym_root + '1']
else:
df1 = futures_hist_prices_dict[sym_root][leg1]
df2 = futures_hist_prices_dict[sym_root][leg2]
df3 = futures_hist_prices_dict[sym_root][leg3]
df0 = generic_futures_hist_prices_dict[sym_root][sym_root + '1']
df = df1 - df2 * 2 + df3
df = pd.concat([df0, df], axis=1)
df = df[lookback_date.date():]
trace1 = go.Scatter(x=df.iloc[:, 0],
y=df.iloc[:, 1],
name=f'{leg1}-{leg2}',
mode='markers')
trace2 = go.Scatter(x=[df.iloc[-1, 0]],
y=[df.iloc[-1, 1]],
name='today',
mode='markers',
marker=dict(color=['red'], size=[20]))
layout_fig = go.Layout(xaxis=dict(title='Generic 1st price'),
yaxis=dict(title='Spread price'),
showlegend=False,
paper_bgcolor='rgba(0,0,0,0)',
plot_bgcolor='rgba(0,0,0,0)')
return go.Figure(data=[trace1, trace2], layout=layout_fig)
def update_historical_time_series_market_commodity_futures_tab1(
rows_data, rows_selected, n_clicks, generic_start_str, generic_end_str,
lookback_window):
# print('historical series called')
sym_root = rows_data[rows_selected[0]]['Contract'][:
-5] # remove e.g. Z2018
# sym = df_config.loc[df_config['Name'] == name]['Quandl Code'].values[0]
lookback_date = convert_date_input(lookback_window, datetime(2008, 1, 1))
if ':' in sym_root:
df = generic_inter_comdty_hist_prices_dict[sym_root]
else:
df = generic_futures_hist_prices_dict[sym_root]
df = df[lookback_date.date():]
generic_start = 1
if (generic_start_str is not None) and (not not generic_start_str):
generic_start = int(generic_start_str)
generic_end = df.shape[1]
if (generic_end_str is not None) and (not not generic_end_str):
generic_end = int(generic_end_str)
traces = [
go.Scatter(x=df[col].index, y=df[col], mode='lines', name=col)
for col in df.columns[generic_start - 1:generic_end]
]
layout_fig = go.Layout(xaxis=dict(title='Date',
rangeslider=dict(visible=False),
type='date'),
yaxis=dict(title='Price'),
legend=dict(orientation="h"),
height=800,
margin=dict(l=0, r=0, t=0, b=0),
paper_bgcolor='rgba(0,0,0,0)',
plot_bgcolor='rgba(0,0,0,0)')
return go.Figure(data=traces, layout=layout_fig)
def update_historical_term_structures_market_commodity_futures_tab1(
rows_data, rows_selected, n_clicks, ione, itwo, ithree, ifour, ifive):
# print('historical term structure called')
sym_root = rows_data[rows_selected[0]]['Contract'][:-5]
if ':' in sym_root:
hist_data = inter_comdty_spread_hist_data_dict[sym_root]
meta_data = inter_comdty_spread_contracts_meta_df[
inter_comdty_spread_contracts_meta_df['Root'] == sym_root]
meta_data.sort_values('Last_Trade_Date', inplace=True)
else:
hist_data = futures_hist_prices_dict[sym_root]
meta_data = futures_contracts_meta_df[futures_contracts_meta_df['Root']
== sym_root]
meta_data.sort_values('Last_Trade_Date', inplace=True)
asofdate = hist_data.index[-1]
s0 = hist_data.loc[asofdate]
s = s0.to_frame()
start_idx = hist_data.shape[0] - 1
if (ione is not None) and (not not ione):
t1 = convert_date_input(ione, datetime.today())
t1 = t1.date()
dateidx1 = hist_data.index.searchsorted(
t1) # first one greater than or equal to
s1 = hist_data.iloc[dateidx1]
s = pd.concat([s, s1], axis=1)
start_idx = min(dateidx1, start_idx)
if (itwo is not None) and (not not itwo):
t2 = convert_date_input(itwo, datetime.today())
t2 = t2.date()
dateidx2 = hist_data.index.searchsorted(
t2) # first one greater than or equal to
s2 = hist_data.iloc[dateidx2]
s = pd.concat([s, s2], axis=1)
start_idx = min(dateidx2, start_idx)
if (ithree is not None) and (not not ithree):
t3 = convert_date_input(ithree, datetime.today())
t3 = t3.date()
dateidx3 = hist_data.index.searchsorted(
t3) # first one greater than or equal to
s3 = hist_data.iloc[dateidx3]
s = pd.concat([s, s3], axis=1)
start_idx = min(dateidx3, start_idx)
if (ifour is not None) and (not not ifour):
t4 = convert_date_input(ifour, datetime.today())
t4 = t4.date()
dateidx4 = hist_data.index.searchsorted(
t4) # first one greater than or equal to
s4 = hist_data.iloc[dateidx4]
s = pd.concat([s, s4], axis=1)
start_idx = min(dateidx4, start_idx)
if (ifive is not None) and (not not ifive):
t5 = convert_date_input(ifive, datetime.today())
t5 = t5.date()
dateidx5 = hist_data.index.searchsorted(
t5) # first one greater than or equal to
s5 = hist_data.iloc[dateidx5]
s = pd.concat([s, s5], axis=1)
start_idx = min(dateidx5, start_idx)
st = s.join(meta_data['Last_Trade_Date'], how='left')
st = st.sort_values('Last_Trade_Date')
# find the first common date
# dateidx_st = st['Last_Trade_Date'].searchsorted(hist_data.index[start_idx])[0]
dateidx_st = st['Last_Trade_Date'].searchsorted(hist_data.index[start_idx])
st = st.iloc[dateidx_st:]
# st.fillna(0.0, inplace=True)
traces = [
go.Scatter(x=st['Last_Trade_Date'],
y=st[c],
name=c.strftime('%Y-%m-%d'),
mode='lines+markers',
hovertext=st.index) for c in st.columns[:-1]
]
layout_fig = go.Layout(title=sym_root,
xaxis={'title': sym_root},
yaxis={'title': 'Price'},
legend=dict(orientation="h"),
paper_bgcolor='rgba(0,0,0,0)',
plot_bgcolor='rgba(0,0,0,0)')
#plotly.offline.plot({'data': traces, 'layout': layout})
return go.Figure(data=traces, layout=layout_fig)
def plot_cross_sectional_market_misc_data(item_selected, ione, itwo, ithree,
ifour, ifive):
df_raw = misc_data_dict[item_selected]
if item_selected in ['USDT']:
df = df_raw
elif item_selected in ['PCR:VIX', 'PCR:SPX', 'PCR:SPY']:
df = pd.concat(
[df_raw['PV'] / df_raw['CV'], df_raw['POI'] / df_raw['COI']],
axis=1)
df.columns = ['PCR:V', 'PCR:OI']
elif 'COT:' in item_selected:
if item_selected not in ['COT:ES', 'COT:NQ', 'COT:UX']:
df = pd.concat([
df_raw['Open Interest:F'],
df_raw['Producer/Merchant/Processor/User:Long:F'] -
df_raw['Producer/Merchant/Processor/User:Short:F'],
df_raw['Swap Dealers:Long:F'] - df_raw['Swap Dealers:Short:F'],
df_raw['Managed Money:Long:F'] -
df_raw['Managed Money:Short:F'],
df_raw['Other Reportables:Long:F'] -
df_raw['Other Reportables:Short:F']
],
axis=1)
df.columns = [
'Open Interest', 'Producers', 'Swap Dealers', 'Managed Money',
'Other Report'
]
df['Commercial'] = df['Producers'] + df['Swap Dealers']
df['Large Spec'] = df['Managed Money'] + df['Other Report']
df['Small Spec'] = 0.0 - df['Commercial'] - df['Large Spec']
else:
df = pd.concat([
df_raw['Open Interest:F'],
df_raw['Dealer Intermediary:Long:F'] -
df_raw['Dealer Intermediary:Short:F'],
df_raw['Asset Manager/Institutional:Long:F'] -
df_raw['Asset Manager/Institutional:Short:F'],
df_raw['Leveraged Funds:Long:F'] -
df_raw['Leveraged Funds:Short:F'],
df_raw['Other Reportables:Long:F'] -
df_raw['Other Reportables:Short:F'],
df_raw['Nonreportable Positions:Long:F'] -
df_raw['Nonreportable Positions:Short:F'],
],
axis=1)
df.columns = [
'Open Interest', 'Dealer Intermediary', 'Asset Manager',
'Leveraged Funds', 'Other Reportables',
'Nonreportable Positions'
]
else:
return None
asofdate = df.index[-1]
s0 = df.loc[asofdate]
s = s0.to_frame()
if (ione is not None) and (not not ione):
t1 = convert_date_input(ione, datetime.today())
t1 = t1.date()
dateidx1 = df.index.searchsorted(
t1) # first one greater than or equal to
s1 = df.iloc[dateidx1]
s = pd.concat([s, s1], axis=1)
if (itwo is not None) and (not not itwo):
t2 = convert_date_input(itwo, datetime.today())
t2 = t2.date()
dateidx2 = df.index.searchsorted(
t2) # first one greater than or equal to
s2 = df.iloc[dateidx2]
s = pd.concat([s, s2], axis=1)
if (ithree is not None) and (not not ithree):
t3 = convert_date_input(ithree, datetime.today())
t3 = t3.date()
dateidx3 = df.index.searchsorted(
t3) # first one greater than or equal to
s3 = df.iloc[dateidx3]
s = pd.concat([s, s3], axis=1)
if (ifour is not None) and (not not ifour):
t4 = convert_date_input(ifour, datetime.today())
t4 = t4.date()
dateidx4 = df.index.searchsorted(
t4) # first one greater than or equal to
s4 = df.iloc[dateidx4]
s = pd.concat([s, s4], axis=1)
if (ifive is not None) and (not not ifive):
t5 = convert_date_input(ifive, datetime.today())
t5 = t5.date()
dateidx5 = df.index.searchsorted(
t5) # first one greater than or equal to
s5 = df.iloc[dateidx5]
s = pd.concat([s, s5], axis=1)
traces = [
go.Scatter(x=s.index,
y=s[c],
name=c.strftime('%Y-%m-%d'),
mode='lines+markers',
hovertext=s.index) for c in s.columns
]
layout_fig = go.Layout(title=item_selected,
xaxis={'title': item_selected},
yaxis={'title': 'Value'},
legend=dict(orientation="h"),
paper_bgcolor='rgba(0,0,0,0)',
plot_bgcolor='rgba(0,0,0,0)')
# plotly.offline.plot({'data': traces, 'layout': layout})
return go.Figure(data=traces, layout=layout_fig)
def plot_time_series_market_misc_data(item_selected, lookback_window):
lookback_date = convert_date_input(lookback_window, datetime(2008, 1, 1))
df_raw = misc_data_dict[item_selected][lookback_date.date():]
if item_selected in ['USDT']:
df = df_raw
elif item_selected in ['PCR:VIX', 'PCR:SPX', 'PCR:SPY']:
df = pd.concat(
[df_raw['PV'] / df_raw['CV'], df_raw['POI'] / df_raw['COI']],
axis=1)
df.columns = ['PCR:V', 'PCR:OI']
elif 'COT:' in item_selected:
if item_selected not in ['COT:ES', 'COT:NQ', 'COT:UX']:
df = pd.concat([
df_raw['Open Interest:F'],
df_raw['Producer/Merchant/Processor/User:Long:F'] -
df_raw['Producer/Merchant/Processor/User:Short:F'],
df_raw['Swap Dealers:Long:F'] - df_raw['Swap Dealers:Short:F'],
df_raw['Managed Money:Long:F'] -
df_raw['Managed Money:Short:F'],
df_raw['Other Reportables:Long:F'] -
df_raw['Other Reportables:Short:F']
],
axis=1)
df.columns = [
'Open Interest', 'Producers', 'Swap Dealers', 'Managed Money',
'Other Report'
]
df['Commercial'] = df['Producers'] + df['Swap Dealers']
df['Large Spec'] = df['Managed Money'] + df['Other Report']
df['Small Spec'] = 0.0 - df['Commercial'] - df['Large Spec']
else:
df = pd.concat([
df_raw['Open Interest:F'],
df_raw['Dealer Intermediary:Long:F'] -
df_raw['Dealer Intermediary:Short:F'],
df_raw['Asset Manager/Institutional:Long:F'] -
df_raw['Asset Manager/Institutional:Short:F'],
df_raw['Leveraged Funds:Long:F'] -
df_raw['Leveraged Funds:Short:F'],
df_raw['Other Reportables:Long:F'] -
df_raw['Other Reportables:Short:F'],
df_raw['Nonreportable Positions:Long:F'] -
df_raw['Nonreportable Positions:Short:F'],
],
axis=1)
df.columns = [
'Open Interest', 'Dealer Intermediary', 'Asset Manager',
'Leveraged Funds', 'Other Reportables',
'Nonreportable Positions'
]
# sym_root = item_selected.split(':')[1]
# hist_price = generic_futures_hist_prices_dict[sym_root][lookback_date.date():].iloc[:,0]
else:
return None
traces = [
go.Scatter(x=df.index, y=df[col], mode='lines', name=col)
for col in df.columns
]
layout_fig = go.Layout(title=item_selected,
xaxis=dict(title=item_selected,
rangeslider=dict(visible=False),
type='date'),
yaxis=dict(title='Value'),
legend=dict(orientation="h"),
height=800,
margin=dict(l=0, r=0, t=0, b=0),
paper_bgcolor='rgba(0,0,0,0)',
plot_bgcolor='rgba(0,0,0,0)')
return go.Figure(data=traces, layout=layout_fig)