from logging import error

from pandas._config.config import reset_option

import streamlit as st

import pandas as pd

import plotly

import matplotlib.pyplot as plt

import bt

import plotly.express as px

from tabulate import tabulate

from functions import alloc_table, balance_table, line_chart, monthly_returns_table, optomize_table, plot_pie, display_stats_combined, results_to_df, highlight_cols, scatter_plot, short_stats_table, stats_table, sum_table, monthly_table

st.set_page_config(layout="wide") #makes page wider

@st.cache

def get_data_f(dontchange):

return data_cache

dontchange = 0

data = get_data_f(dontchange)

# st.markdown(

# """

# <style>

# .main {

# background-color: #478f7c;

# }

# </style>

# """,

# unsafe_allow_html=True

# )

word = "Dashboard"

#st.title(word)

class WeighSpecified(bt.Algo):

return True

st.sidebar.write("Options")

option = st.sidebar.selectbox("Select an Option", ('Portfolio Optimizer','Flexible Dashboard', 'BTC Portfolio Dashboard'))

start_date = '2017-01-01'

if ( option == 'Chart'):

#Beta Columns and Containers

col1_s, col2_s = st.sidebar.beta_columns(2)

col1, col2 = st.beta_columns((2, 1))

col1_header = col1.beta_container()

col2_header = col2.beta_container()

col1_graph = col1.beta_container()

col2_graph = col2.beta_container()

col1_second = col1.beta_container()

col2_second = col2.beta_container()

#Sidebar Inputs

stock_choice_1 = col1_s.selectbox( "Ticker 1", ('spy', 'efa', 'iwm', 'vwo', 'ibb', 'agg', 'hyg', 'gld', 'slv', 'tsla', 'aapl', 'msft', 'qqq', 'btc-usd', 'eth-usd')) #get ticker

percent_1 = col2_s.text_input( "% Allocation", value = 55, max_chars= 3, ) # get percent

stock_choice_1 = stock_choice_1.lower() #bt likes lower case

data_1 = bt.get(stock_choice_1, start = start_date) # get the data

stock_choice_2 = col1_s.selectbox( "Ticker 2", ('agg', 'spy', 'efa', 'iwm', 'vwo', 'ibb', 'hyg', 'gld', 'slv', 'tsla', 'aapl', 'msft', 'qqq', 'btc-usd', 'eth-usd'))

percent_2 = col2_s.text_input( "% Allocation", value = 40, max_chars= 3)

stock_choice_2 = stock_choice_2.lower()

data_2 = bt.get(stock_choice_2, start = start_date)

stock_choice_3 = col1_s.selectbox( "Ticker 3", ('btc-usd', 'spy', 'efa', 'iwm', 'vwo', 'ibb', 'agg', 'hyg', 'gld', 'slv', 'tsla', 'aapl', 'msft', 'qqq', 'eth-usd'))

percent_3 = col2_s.text_input( "% Allocation", value = 5, max_chars= 3)

stock_choice_3 = stock_choice_3.lower()

data_3 = bt.get(stock_choice_3, start = start_date)

#allows us to combine the datasets to account for the difference in reg vs. Crypto

data = data_1.join(data_2, how='outer')

data = data.join(data_3, how= 'outer')

data = data.dropna()

#need the '-' in cryptos to get the data, but bt needs it gone to work

stock_choice_1 = stock_choice_1.replace('-', '')

stock_choice_2 = stock_choice_2.replace('-', '')

stock_choice_3 = stock_choice_3.replace('-', '')

#Buttons

rebalances = col1_graph.selectbox("Rebalancing Timeline", ('Daily', 'Monthly', 'Yearly', 'None'))

#creating Strategy and Backtest

stock_list = stock_choice_1 +',' + stock_choice_2 + ',' + stock_choice_3 #list of tickers to get data for

stock_list_plt = [stock_choice_1, stock_choice_2, stock_choice_3]

percent_list = [percent_1, percent_2, percent_3]

stock_dic = {stock_choice_1: float(percent_1)/100, stock_choice_2: float(percent_2)/100, stock_choice_3: float(percent_3)/100} #dictonary for strat

stock_dic_control = {'spy': float(60)/100, 'agg': float(40)/100, stock_choice_3: float(0)/100}

stock_dic_spy = {'spy': float(100)/100, 'agg': float(0)/100, stock_choice_3: float(0)/100}

stock_dic_agg = {'spy': float(0)/100, 'agg': float(100)/100, stock_choice_3: float(0)/100}

strategy_ = bt.Strategy('Your Strategy Monthly',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

strategy_spy = bt.Strategy('SPY',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_spy),

bt.algos.Rebalance()]) #Creating strategy

strategy_agg = bt.Strategy('AGG',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_agg),

bt.algos.Rebalance()]) #Creating strategy

if (rebalances == 'Daily'):

strategy_ = bt.Strategy('Your Strategy Daily',

[bt.algos.RunDaily(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 Daily',

[bt.algos.RunDaily(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

elif (rebalances == 'Monthly'):

strategy_ = bt.Strategy('Your Strategy Monthly',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 Monthly',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

elif (rebalances == 'Yearly'):

strategy_ = bt.Strategy('Your Strategy Yearly',

[bt.algos.RunYearly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 Yearly',

[bt.algos.RunYearly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

elif (rebalances == 'None'):

strategy_ = bt.Strategy('Your Strategy None',

[bt.algos.RunOnce(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 None',

[bt.algos.RunOnce(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

test_control = bt.Backtest(strategy_control, data)

results_control = bt.run(test_control)

test_spy = bt.Backtest(strategy_spy, data)

results_spy = bt.run(test_spy)

test_agg = bt.Backtest(strategy_agg, data)

results_agg = bt.run(test_agg)

test = bt.Backtest(strategy_, data)

results = bt.run(test)

#Line Chart

ser = results._get_series(None).rebase() #gets all the daily balances as a series

ser2 = results_control._get_series(None).rebase()

result_final = pd.concat([ser, ser2], axis=1) #makes dataframe for both series

col1_header.header("Returns Graph")

col1_graph.line_chart(result_final)

#Pie Chart

if (rebalances == 'None'): #pie chart is wrong since no rebalances

key = results._get_backtest(0) #Chunk of code is how to get the weights, straight from doc

filter = None

if filter is not None:

data = results.backtests[key].security_weights[filter]

else:

data = results.backtests[key].security_weights

for i in range(len(percent_list)): #puts all the right values into the percent list

percent_list[i] = str(round(data[stock_list_plt[i]].iloc[-1]*100))

fig = plot_pie(stock_list_plt, percent_list)

col2_header.header("Pie Chart")

col2_header.pyplot(fig)

#Display Results

results_list = [results, results_control, results_spy, results_agg] #list of results objects

results_df = results_to_df(results_list) #list of the results but now in dataframe

stats = display_stats_combined(results_list)

if (col2_second.button("Display Stats")):#button logic

if(col2_second.button("Hide Stats")):

do_nothing = 0 #literally do nothing

col2_second.dataframe(stats)

col1_second.write(results.display_lookback_returns()) # displays the shortened stats

#Display the Monthly Returns

mon_table = monthly_returns_table(results_list)

st.dataframe(mon_table)

#Scatter of Risk vs Return

fig = scatter_plot(results_df) #scatter function in functions

col2_second.plotly_chart(fig)

#Allocation Table

rebalance_list = [3, 4, 5]

fig = alloc_table(stock_list_plt, percent_list, rebalance_list)

col2.plotly_chart(fig)

elif ( option == 'BTC Portfolio Dashboard'):

#Beta Columns

col1_s, col2_s = st.sidebar.beta_columns(2)

col1, col2 = st.beta_columns((1, 2))

col3, col4, col5 = st.beta_columns((1,1,3))

col1_top = col1.beta_container()

col1_middle = col1.beta_container()

col1_bot = col1.beta_container()

col2t =col2.beta_container()

col2b =col2.beta_container()

table = st.beta_container()

#Creating Strategy and Backtest

slider_input = col1_middle.slider('Percent of BTC-USD in Portfolio', min_value= 0, max_value= 10, value= 5 )

#hardcoding in the values since we dont have user input

stock_choice_1 = 'spy'

stock_choice_2 = 'agg'

stock_choice_3 = 'btc-usd'

stock_list_plt = [stock_choice_1, stock_choice_2, stock_choice_3]

percent_1 = 60-slider_input

percent_2 = 40

percent_3 = slider_input

percent_list = [percent_1, percent_2, percent_3]

#get data seperatly because crypto and reg data dont work together

# data_1 = bt.get(stock_choice_1, start = start_date)

# data_2 = bt.get(stock_choice_2, start = start_date)

# data_3 = bt.get(stock_choice_3, start = start_date)

#Allows for crypto and stock to be in a dataframe

# data = data_1.join(data_2, how='outer')

# data = data.join(data_3, how= 'outer')

# data = data.dropna()

stock_choice_3 = stock_choice_3.replace('-', '') #get data with btc-usd but then bt likes btcusd

stock_dic = {stock_choice_1: float(percent_1)/100, stock_choice_2: float(percent_2)/100, stock_choice_3: float(percent_3)/100} #dictonary for strat

stock_dic_control = {stock_choice_1: float(60)/100, stock_choice_2: float(40)/100, stock_choice_3: float(0)/100} #60-40

stock_dic_spy = {'spy': float(100)/100, 'agg': float(0)/100, stock_choice_3: float(0)/100} #all spy

stock_dic_agg = {'spy': float(0)/100, 'agg': float(100)/100, stock_choice_3: float(0)/100} #all agg

strategy_ = bt.Strategy('Your Strategy',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 Portfolio',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

strategy_spy = bt.Strategy('SPY',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_spy),

bt.algos.Rebalance()]) #Creating strategy

strategy_agg = bt.Strategy('AGG',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_agg),

bt.algos.Rebalance()]) #Creating strategy

test_control = bt.Backtest(strategy_control, data)

results_control = bt.run(test_control)

test_spy = bt.Backtest(strategy_spy, data)

results_spy = bt.run(test_spy)

test_agg = bt.Backtest(strategy_agg, data)

results_agg = bt.run(test_agg)

test = bt.Backtest(strategy_, data)

results = bt.run(test)

results_list = [results, results_control, results_spy, results_agg]

results_df = results_to_df(results_list) #list of the results but now in dataframe

#Line Chart

fig = line_chart(results_list)

col1.header("Returns Graph")

col2.plotly_chart(fig)

col5.write("- Click on the legend entries to choose which datasets to display")

#Pie Chart

fig = plot_pie(stock_list_plt, percent_list)

col1_top.header("Pie Chart")

col1_top.pyplot(fig)

#Display Results

# results_list = [results, results_control, results_spy, results_agg] #list of results objects

# results_df = results_to_df(results_list) #list of the results but now in dataframe

# stats = display_stats_combined(results_list)

# if (col2_second.button("Display Stats")):

# if(col2_second.button("Hide Stats")):

# yo = 1

# col2_second.dataframe(stats)

# col1_second.write(results.display_lookback_returns())

#Display the Monthly Returns

mon_table = monthly_returns_table(results_list)

#st.dataframe(mon_table.style.apply(highlight_cols, axis = None))

#Scatter of Risk vs Return

fig = scatter_plot(results_df) #scatter function in functions

fig.update_layout(width = 750, height = 500)

col2t.plotly_chart(fig, width = 750, height =500)

#Allocation Table

rebalance_list = [3, 4, 5]

fig = alloc_table(stock_list_plt, percent_list, rebalance_list)

fig.update_layout(width = 400, height = 130)

#col1.plotly_chart(fig, width = 400, height = 130)

#Balance Table

fig = balance_table(results, results_control)

fig.update_layout(width = 380, height = 75)

col1.plotly_chart(fig, width = 380, height = 75)

#Short Stats Table

fig = short_stats_table(results_list)

fig.update_layout(width = 380, height = 300)

col1.header("Return Statistics")

col1.plotly_chart(fig, width = 380, height = 300)

#Monthly Table

#my_expander = st.beta_expander("Show Monthly Returns")

fig = monthly_table(results_list)

fig.update_layout(width = 1100, height = 2000)

st.plotly_chart(fig, width = 1100, height = 2000)

#Stats Table

stats_expander = col1.beta_expander("Click to Show Strategy Statistics")

fig = stats_table(results_list)

fig.update_layout(width = 380)

stats_expander.plotly_chart(fig, width = 380)

if ( option == 'Flexible Dashboard'):

#Beta Columns and Containers

col1_s, col2_s = st.sidebar.beta_columns(2)

col1, col2 = st.beta_columns((1, 2))

col3, col4, col5 = st.beta_columns((1,1,3))

col1_top = col1.beta_container()

col1_middle = col1.beta_container()

col1_bot = col1.beta_container()

col2t =col2.beta_container()

col2b =col2.beta_container()

col1_header = col1.beta_container()

col2_header = col2.beta_container()

col1_graph = col1.beta_container()

col2_graph = col2.beta_container()

col1_second = col1.beta_container()

col2_second = col2.beta_container()

table = st.beta_container()

#Sidebar Inputs

stock_choice_1 = col1_s.selectbox( "Ticker 1", ('spy', 'efa', 'iwm', 'vwo', 'ibb', 'agg', 'hyg', 'gld', 'slv', 'tsla', 'aapl', 'msft', 'qqq', 'btc-usd', 'eth-usd')) #get ticker

percent_1 = col2_s.text_input( "% Allocation", value = 55, max_chars= 3, ) # get percent

stock_choice_1 = stock_choice_1.lower() #bt likes lower case

#data_1 = bt.get(stock_choice_1, start = start_date) # get the data

stock_choice_2 = col1_s.selectbox( "Ticker 2", ('agg', 'spy', 'efa', 'iwm', 'vwo', 'ibb', 'hyg', 'gld', 'slv', 'tsla', 'aapl', 'msft', 'qqq', 'btc-usd', 'eth-usd'))

percent_2 = col2_s.text_input( "% Allocation", value = 40, max_chars= 3)

stock_choice_2 = stock_choice_2.lower()

#data_2 = bt.get(stock_choice_2, start = start_date)

stock_choice_3 = col1_s.selectbox( "Ticker 3", ('btc-usd', 'spy', 'efa', 'iwm', 'vwo', 'ibb', 'agg', 'hyg', 'gld', 'slv', 'tsla', 'aapl', 'msft', 'qqq', 'eth-usd'))

percent_3 = col2_s.text_input( "% Allocation", value = 5, max_chars= 3)

stock_choice_3 = stock_choice_3.lower()

#data_3 = bt.get(stock_choice_3, start = start_date)

if(float(percent_1)+float(percent_2)+float(percent_3) != 100):

st.sidebar.error("Allocation Must Equal 100")

#allows us to combine the datasets to account for the difference in reg vs. Crypto

# data = data_1.join(data_2, how='outer')

# data = data.join(data_3, how= 'outer')

# data = data.dropna()

#data_con = bt.get('spy,agg,gme', start = start_date)

#need the '-' in cryptos to get the data, but bt needs it gone to work

stock_choice_1 = stock_choice_1.replace('-', '')

stock_choice_2 = stock_choice_2.replace('-', '')

stock_choice_3 = stock_choice_3.replace('-', '')

#Buttons

#rebalances = col1_graph.selectbox("Rebalancing Timeline", ('Daily', 'Monthly', 'Yearly', 'None'))

rebalances = 'Monthly'

#creating Strategy and Backtest

stock_list = stock_choice_1 +',' + stock_choice_2 + ',' + stock_choice_3 #list of tickers to get data for

stock_list_plt = [stock_choice_1, stock_choice_2, stock_choice_3]

percent_list = [percent_1, percent_2, percent_3]

stock_dic = {stock_choice_1: float(percent_1)/100, stock_choice_2: float(percent_2)/100, stock_choice_3: float(percent_3)/100} #dictonary for strat

stock_dic_control = {'spy': float(60)/100, 'agg': float(40)/100, stock_choice_3: float(0)/100}

stock_dic_spy = {'spy': float(100)/100, 'agg': float(0)/100, stock_choice_3: float(0)/100}

stock_dic_agg = {'spy': float(0)/100, 'agg': float(100)/100, stock_choice_3: float(0)/100}

strategy_ = bt.Strategy('Your Strategy',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 Portfolio',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

strategy_spy = bt.Strategy('SPY',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_spy),

bt.algos.Rebalance()]) #Creating strategy

strategy_agg = bt.Strategy('AGG',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_agg),

bt.algos.Rebalance()]) #Creating strategy

strategy_daily = bt.Strategy('Daily',

[bt.algos.RunDaily(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_monthly = bt.Strategy('Monthly',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_yearly = bt.Strategy('Yearly',

[bt.algos.RunYearly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_none = bt.Strategy('No Rebalances',

[bt.algos.RunOnce(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

#old rebalances

if (rebalances == 'Daily'):

strategy_daily = bt.Strategy('Your Strategy Daily',

[bt.algos.RunDaily(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 Daily',

[bt.algos.RunDaily(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

elif (rebalances == 'Monthly'):

strategy_monthly = bt.Strategy('Monthly',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 Portfolio',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

elif (rebalances == 'Yearly'):

strategy_yearly = bt.Strategy('Your Strategy Yearly',

[bt.algos.RunYearly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 Yearly',

[bt.algos.RunYearly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

elif (rebalances == 'None'):

strategy_none = bt.Strategy('Your Strategy None',

[bt.algos.RunOnce(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_control = bt.Strategy('60-40 None',

[bt.algos.RunOnce(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_control),

bt.algos.Rebalance()]) #Creating strategy

#results

test_control = bt.Backtest(strategy_control, data)

results_control = bt.run(test_control)

test_spy = bt.Backtest(strategy_spy, data)

results_spy = bt.run(test_spy)

test_agg = bt.Backtest(strategy_agg, data)

results_agg = bt.run(test_agg)

test = bt.Backtest(strategy_, data)

results = bt.run(test)

#Rebalance Strategies

test_d = bt.Backtest(strategy_daily, data)

results_daily = bt.run(test_d)

test_m = bt.Backtest(strategy_monthly, data)

results_monthly = bt.run(test_m)

test_y = bt.Backtest(strategy_yearly, data)

results_yearly = bt.run(test_y)

test_n = bt.Backtest(strategy_none, data)

results_none = bt.run(test_n)

results_list = [results, results_control, results_spy, results_agg] #list of results objects

results_list_reb = [results_daily, results_monthly, results_yearly, results_none]

#Line Chart

fig = line_chart(results_list)

fig.update_layout(width = 700)

fig2 = line_chart(results_list_reb)

fig2.update_layout(width = 700)

figure = fig

box = col2b.checkbox('Compare Rebalancing Options for Your Strategy')

if box:

figure = fig2

col2b.plotly_chart(figure)

col5.write("- Click on the legend entries to choose which datasets to display")

#Pie Chart

if (rebalances == 'None'): #pie chart is wrong since no rebalances

key = results._get_backtest(0) #Chunk of code is how to get the weights, straight from doc

filter = None

if filter is not None:

data = results.backtests[key].security_weights[filter]

else:

data = results.backtests[key].security_weights

for i in range(len(percent_list)): #puts all the right values into the percent list

percent_list[i] = str(round(data[stock_list_plt[i]].iloc[-1]*100))

fig = plot_pie(stock_list_plt, percent_list)

fig.set_facecolor('#fafafa')

col1_top.header("PORTFOLIO ALLOCATION")

col1_top.pyplot(fig)

#Display Results

results_df = results_to_df(results_list) #list of the results but now in dataframe

stats = display_stats_combined(results_list)

# if (col2_second.button("Display Stats")):#button logic

# if(col2_second.button("Hide Stats")):

# do_nothing = 0 #literally do nothing

#col2_second.dataframe(stats)

#col1_second.write(results.display_lookback_returns()) # displays the shortened stats

#Display the Monthly Returns

mon_table = monthly_returns_table(results_list)

#st.dataframe(mon_table.style.apply(highlight_cols, axis = None))

#Scatter of Risk vs Return

fig = scatter_plot(results_df) #scatter function in functions

fig.update_layout(width = 750, height = 500)

col2t.plotly_chart(fig, width = 750, height =500)

#Allocation Table

rebalance_list = [3, 4, 5]

fig = alloc_table(stock_list_plt, percent_list, rebalance_list)

fig.update_layout(width = 400, height = 130)

#col1.plotly_chart(fig, width = 400, height = 130)

#Balance Table

fig = balance_table(results, results_control)

fig.update_layout(width = 400, height = 75)

col1.plotly_chart(fig, width = 400, height = 75)

#Short Stats Table

fig = short_stats_table(results_list)

fig.update_layout(width = 380, height = 300)

col1.header("Return Statistics")

col1.plotly_chart(fig, width = 380, height = 300)

#Monthly Table

#my_expander = st.beta_expander("Show Monthly Returns")

fig = monthly_table(results_list)

fig.update_layout(width = 1100, height = 800)

st.plotly_chart(fig, width = 1100, height = 800)

#Stats Table

stats_expander = col1.beta_expander("Click to Show Strategy Statistics")

fig = stats_table(results_list)

fig.update_layout(width = 380)

stats_expander.plotly_chart(fig, width = 380)

elif (option == 'Portfolio Optimizer'):

#Beta Columns

col1_s, col2_s = st.sidebar.beta_columns(2)

col1, col2 = st.beta_columns((1, 2))

#Get data

stock_symbols = st.sidebar.text_input("Enter the Stock Tickers Spaced", value = "spy iwm eem efa gld agg hyg" )

crypto_symbols = st.sidebar.text_input("Enter Crypto Tickers Spaced", value= 'btc-usd')

stock_symbols = stock_symbols.replace(' ', ',')

crypto_symbols = crypto_symbols.replace(' ', ',')

data_type = st.sidebar.selectbox("Select the Data Frequency", ('Daily Data', 'Monthly Data', 'Quarterly Data', 'Yearly Data'))

# symbols = 'spy,iwm,eem,efa,gld,agg,hyg'

# crypto_symbols = 'btc-usd,eth-usd'

stock_data = bt.get(stock_symbols, start='1993-01-01')

crypto_data = bt.get(crypto_symbols, start='2016-01-01')

#Merge into dataframe

data_ = crypto_data.join(stock_data, how='outer')

data_ = data_.dropna()

#Daily optimal

if (data_type == "Daily Data"):

#gets daily optimal data

returns = data_.to_log_returns().dropna()

daily_opt = returns.calc_mean_var_weights().as_format(".2%")

#table

fig = optomize_table(daily_opt)

col1.header("Daily Data")

fig.update_layout(width = 300, height = 450)

col1.plotly_chart(fig, width = 300, height = 450)

#preparing data for charts

stock_dic = daily_opt.to_dict()

for key in stock_dic: #makes percents numbers

stock_dic[key] = float(stock_dic[key].replace('%', ''))

stock_dic[key] = stock_dic[key]/100

stock_list = list(stock_dic.keys()) #convert the dictionary into lists for plotting

percent_list = list(stock_dic.values())

temp = []

temp_stock = []

for i in range(len(percent_list)): #Takes out values of 0

if (percent_list[i] != 0):

temp.append(percent_list[i])

temp_stock.append(stock_list[i])

stock_list = temp_stock

percent_list= temp

strategy_color = '#A90BFE'

P6040_color = '#FF7052'

spy_color = '#66F3EC'

agg_color = '#67F9AF'

stock_dic_spy = {'spy': float(100)/100, 'agg': float(0)/100, 'gme': float(0)/100} #all spy

stock_dic_agg = {'spy': float(0)/100, 'agg': float(100)/100, 'gme': float(0)/100} #all agg

strategy_op = bt.Strategy('Your Portolio Optomized',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_port = bt.Strategy('Your Portolio Equal',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighEqually(),

bt.algos.Rebalance()]) #Creating strategy

strategy_spy = bt.Strategy('SPY',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_spy),

bt.algos.Rebalance()]) #Creating strategy

strategy_agg = bt.Strategy('AGG',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_agg),

bt.algos.Rebalance()]) #Creating strategy

test_op = bt.Backtest(strategy_op, data_)

results_op = bt.run(test_op)

test_port = bt.Backtest(strategy_port, data_)

results_port = bt.run(test_port)

test_spy = bt.Backtest(strategy_spy, data)

results_spy = bt.run(test_spy)

test_agg = bt.Backtest(strategy_agg, data)

results_agg = bt.run(test_agg)

#pie chart

results_list = [results_op, results_port]

pie_colors = [strategy_color, P6040_color, spy_color, agg_color, '#7496F3', '#B7FA59', 'brown', '#EE4444', 'gold']

fig = plot_pie(stock_list, percent_list, pie_colors)

#fig.set_size_inches(18.5, 18.5, forward=True) #how to change dimensions since pie is in matplotlib

#col1.header("Optomized Portfolio")

col1.pyplot(fig)

#line chart

fig = line_chart(results_list)

fig.update_layout(width = 750, height = 400)

col2.header("Daily Performance")

col2.plotly_chart(fig, width = 750, height = 400)

#Scatter PLot

results_list.append(results_spy)

results_list.append(results_agg)

results_df = results_to_df(results_list)

fig = scatter_plot(results_df) #scatter function in functions

fig.update_layout(width = 750, height = 500)

col2.plotly_chart(fig, width = 750, height =500)

#Monthly optimal

if (data_type == "Monthly Data"):

#gets daily optimal data

returns = data_.asfreq("M",method='ffill').to_log_returns().dropna()

mon_opt = returns.calc_mean_var_weights().as_format(".2%")

#table

fig = optomize_table(mon_opt)

col1.header("Monthly Data")

fig.update_layout(width = 300, height = 450)

col1.plotly_chart(fig, width = 300, height = 450)

#preparing data for charts

stock_dic = mon_opt.to_dict()

for key in stock_dic: #makes percents numbers

stock_dic[key] = float(stock_dic[key].replace('%', ''))

stock_dic[key] = stock_dic[key]/100

stock_list = list(stock_dic.keys()) #convert the dictionary into lists for plotting

percent_list = list(stock_dic.values())

temp = []

temp_stock = []

for i in range(len(percent_list)): #Takes out values of 0

if (percent_list[i] != 0):

temp.append(percent_list[i])

temp_stock.append(stock_list[i])

stock_list = temp_stock

percent_list= temp

strategy_color = '#A90BFE'

P6040_color = '#FF7052'

spy_color = '#66F3EC'

agg_color = '#67F9AF'

stock_dic_spy = {'spy': float(100)/100, 'agg': float(0)/100, 'gme': float(0)/100} #all spy

stock_dic_agg = {'spy': float(0)/100, 'agg': float(100)/100, 'gme': float(0)/100} #all agg

strategy_op = bt.Strategy('Your Portolio Optomized',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_port = bt.Strategy('Your Portolio Equal',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighEqually(),

bt.algos.Rebalance()]) #Creating strategy

strategy_spy = bt.Strategy('SPY',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_spy),

bt.algos.Rebalance()]) #Creating strategy

strategy_agg = bt.Strategy('AGG',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_agg),

bt.algos.Rebalance()]) #Creating strategy

test_op = bt.Backtest(strategy_op, data_)

results_op = bt.run(test_op)

test_port = bt.Backtest(strategy_port, data_)

results_port = bt.run(test_port)

test_spy = bt.Backtest(strategy_spy, data)

results_spy = bt.run(test_spy)

test_agg = bt.Backtest(strategy_agg, data)

results_agg = bt.run(test_agg)

#pie chart

results_list = [results_op, results_port]

pie_colors = [strategy_color, P6040_color, spy_color, agg_color, '#7496F3', '#B7FA59', 'brown', '#EE4444', 'gold']

fig = plot_pie(stock_list, percent_list, pie_colors)

#fig.set_size_inches(18.5, 18.5, forward=True) #how to change dimensions since pie is in matplotlib

#col1.header("Optomized Portfolio")

col1.pyplot(fig)

#line chart

fig = line_chart(results_list)

fig.update_layout(width = 750, height = 400)

col2.header("Daily Performance")

col2.plotly_chart(fig, width = 750, height = 400)

#Scatter PLot

results_list.append(results_spy)

results_list.append(results_agg)

results_df = results_to_df(results_list)

fig = scatter_plot(results_df) #scatter function in functions

fig.update_layout(width = 750, height = 500)

col2.plotly_chart(fig, width = 750, height =500)

#Quarterly Optimal

if (data_type == "Quarterly Data"):

#gets quarterly optimal data

quarterly_rets = data_.asfreq("Q",method='ffill').to_log_returns().dropna()

quart_opt = quarterly_rets.calc_mean_var_weights().as_format(".2%")

#table

fig = optomize_table(quart_opt)

col1.header("Quarterly Data")

fig.update_layout(width = 300, height = 450)

col1.plotly_chart(fig, width = 300, height = 450)

#preparing data for charts

stock_dic = quart_opt.to_dict()

for key in stock_dic: #makes percents numbers

stock_dic[key] = float(stock_dic[key].replace('%', ''))

stock_dic[key] = stock_dic[key]/100

stock_list = list(stock_dic.keys()) #convert the dictionary into lists for plotting

percent_list = list(stock_dic.values())

temp = []

temp_stock = []

for i in range(len(percent_list)): #Takes out values of 0

if (percent_list[i] != 0):

temp.append(percent_list[i])

temp_stock.append(stock_list[i])

stock_list = temp_stock

percent_list= temp

strategy_color = '#A90BFE'

P6040_color = '#FF7052'

spy_color = '#66F3EC'

agg_color = '#67F9AF'

stock_dic_spy = {'spy': float(100)/100, 'agg': float(0)/100, 'gme': float(0)/100} #all spy

stock_dic_agg = {'spy': float(0)/100, 'agg': float(100)/100, 'gme': float(0)/100} #all agg

strategy_op = bt.Strategy('Your Portolio Optomized',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic),

bt.algos.Rebalance()]) #Creating strategy

strategy_port = bt.Strategy('Your Portolio Equal',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighEqually(),

bt.algos.Rebalance()]) #Creating strategy

strategy_spy = bt.Strategy('SPY',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_spy),

bt.algos.Rebalance()]) #Creating strategy

strategy_agg = bt.Strategy('AGG',

[bt.algos.RunMonthly(),

bt.algos.SelectAll(),

bt.algos.WeighSpecified(**stock_dic_agg),

bt.algos.Rebalance()]) #Creating strategy

test_op = bt.Backtest(strategy_op, data_)

results_op = bt.run(test_op)

test_port = bt.Backtest(strategy_port, data_)

results_port = bt.run(test_port)

test_spy = bt.Backtest(strategy_spy, data)

results_spy = bt.run(test_spy)

test_agg = bt.Backtest(strategy_agg, data)

results_agg = bt.run(test_agg)

#pie chart

results_list = [results_op, results_port]

pie_colors = [strategy_color, P6040_color, spy_color, agg_color, '#7496F3', '#B7FA59', 'brown', '#EE4444', 'gold']

fig = plot_pie(stock_list, percent_list, pie_colors)

#fig.set_size_inches(18.5, 18.5, forward=True) #how to change dimensions since pie is in matplotlib

#col1.header("Optomized Portfolio")

col1.pyplot(fig)

#line chart

fig = line_chart(results_list)

fig.update_layout(width = 750, height = 400)

col2.header("Daily Performance")

col2.plotly_chart(fig, width = 750, height = 400)

#Scatter PLot

results_list.append(results_spy)

results_list.append(results_agg)

results_df = results_to_df(results_list)

fig = scatter_plot(results_df) #scatter function in functions

fig.update_layout(width = 750, height = 500)

col2.plotly_chart(fig, width = 750, height =500)

#Yearly Optimal

if (data_type == "Yearly Data"):

#gets Yearly optimal data

year_rets = data_.asfreq("Y",method='ffill').to_log_returns().dropna()

year_opt = year_rets.calc_mean_var_weights().as_format(".2%")

#table

fig = optomize_table(year_opt)

col1.header("Yearly Data")

fig.update_layout(width = 300, height = 450)

col1.plotly_chart(fig, width = 300, height = 450)

#preparing data for charts

stock_dic = year_opt.to_dict()