import pandas as pd import seaborn as sns df_obs = pd.read_csv('/work/20170103_PFE_nasdaq_1sec.csv') # df_obm = pd.read_csv('/work/20170103_PFE_nasdaq_1min.csv') df_obs['Mid_Price'] = (df_obs['Bid_Price'] + df_obs['Offer_Price']) / 2 df_obs['Volume_Weighted_Mid_Price'] = (df_obs['Bid_Price'] * df_obs['Offer_Size']) + (df_obs['Offer_Price'] * df_obs['Bid_Size']) / (df_obs['Offer_Size'] + df_obs['Bid_Size']) df_obs['return'] = df_obs['Mid_Price'].diff() df_obs['fractional_return'] = df_obs['Mid_Price'].diff() / df_obs['Mid_Price'] df_obs

df_obs.isnull().sum()

df_obs.dropna(how='any', inplace=True) df_obs.isnull().sum()

(df_obs == 0).sum()

import plotly.graph_objects as go # Create traces fig = go.Figure() fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['Mid_Price'], mode='lines', name='lines')) fig.update_layout( title="Mid Price", xaxis_title="Datetime", yaxis_title="Mid Price" ) fig.show() fig = go.Figure() fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['Volume_Weighted_Mid_Price'], mode='lines', name='lines')) fig.update_layout( title="Volume Weighted Mid Price", xaxis_title="Datetime", yaxis_title="WVMP" ) fig.show() fig = go.Figure() fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['return'], mode='lines', name='lines')) fig.update_layout( title="Return", xaxis_title="Datetime", yaxis_title="Return" ) fig.show() fig = go.Figure() fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['fractional_return'], mode='lines', name='lines')) fig.update_layout( title="Fractional", xaxis_title="Datetime", yaxis_title="frational_return" ) fig.show()

df_obs['Volatility_60s'] = df_obs['return'].rolling(window = 60).std() df_obs['Volatility_3600s'] = df_obs['return'].rolling(window = 3600).std()

from plotly.subplots import make_subplots # Create figure with secondary y-axis fig = make_subplots(specs=[[{"secondary_y": True}]]) fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['Volatility_60s'], mode='lines', name='Volatility 1 minute')) fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['Volatility_3600s'], mode='lines', name='Volatility 1 hour')) fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['return'], mode='lines', name='Return'), secondary_y=True) fig.update_layout( title="Volatility", xaxis_title="Datetime", yaxis_title="Volatility" ) fig.show()

df_obs

df_trades = pd.read_csv('/work/20170103_PFE_trades_1sec.csv') df_trades

df_trades.isnull().sum()

df_trades.dropna(how='any', inplace=True)

import plotly.graph_objects as go from plotly.subplots import make_subplots fig = go.Figure() fig.add_trace(go.Scatter(x=df_trades['DateTime'], y=df_trades['Trade Price'], mode='lines', name='lines')) fig.update_layout( title="Trade Price", xaxis_title="Time", yaxis_title="Trade Price" ) fig.show() fig = make_subplots(specs=[[{"secondary_y": True}]]) fig.add_trace(go.Scatter(x=df_trades['DateTime'], y=df_trades['Trade Volume'], mode='lines', name='Volume')) fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['Volatility_60s'], mode='lines', name='Volatility 1 minute'), secondary_y=True) fig.add_trace(go.Scatter(x=df_obs['DateTime'], y=df_obs['Volatility_3600s'], mode='lines', name='Volatility 1 hour'), secondary_y=True) fig.update_layout( title="Trade Volume", xaxis_title="Time", yaxis_title="Trade Volume" ) fig.show()

df_obs[['DateTime','Volatility_3600s']].plot.line()