import numpy as np
import pandas as pd
import os
import openpyxl
import csv
from scipy.optimize import minimize
from tkinter import *

stocksymbols = []

#begin gui

window = Tk()
window.title("Portfolio Entry")
window.configure(background="black")

possibles = (pd.read_json('https://api.iextrading.com/1.0/ref-data/symbols').loc[:, ("symbol")]).tolist()

Label (window, text="Enter a ticker: ", bg="black", fg="white", font="none 12 bold") .grid(row=1, column=0, sticky=W)

textentry = Entry(window, width=12, bg="white")
textentry.grid(row=2, column=0, sticky=W)

def addsymbol():
    entered_text = textentry.get().upper()
    if entered_text in possibles:
        stocksymbols.append(entered_text)
        textentry.delete(0, END)
        portfolio.delete(0, END)
        portfolio.insert(0, stocksymbols)
    else:
        print("\n Sorry, invalid ticker \n")

def removesymbol():
    stocksymbols.pop()
    portfolio.delete(0, END)
    portfolio.insert(0, stocksymbols)

Button(window, text="SUBMIT SYMBOL", width=13, command=addsymbol) .grid(row=3, column=0, sticky=W)
Button(window, text="REMOVE SYMBOL", width=13, command=removesymbol) .grid(row=3, column=1, sticky=W)

Label (window, text="Your portfolio: ", bg="black", fg="white", font="none 12 bold") . grid(row=5, column=0, sticky=W)
portfolio = Entry(window, width=50, bg="white")
portfolio.grid(row=6, column=0, sticky=W)


window.mainloop()#end gui and continue script with created portfolio




thebigdict = {'Ticker': stocksymbols}
dfbig = pd.DataFrame(thebigdict)
dfbig.rename(index=str)
print(dfbig)

#create list of returns and list of standard deviations

returnslist=[]
standdevlist=[]
for symbolmain in stocksymbols:
    dfmain = pd.DataFrame()
    try:
        dfmain = pd.read_json('https://api.iextrading.com/1.0/stock/'+symbolmain+'/chart/5y')
    except Exception: pass #some symbols don't have json urls
    try:
        dailyreturnavgmain = '{:f}'.format((dfmain.changePercent-(1.76/365)).mean())
        standdevmain = '{:f}'.format((dfmain.changePercent-(1.76/365)).std())
    except AttributeError: pass #some symbol jsons don't have changePercent attributes
    returnslist.append(float(dailyreturnavgmain))
    standdevlist.append(float(standdevmain))

    correlationlist=[]#for the correlations of the symbol in stocksymbols with all other symbols
    for symbolsecond in stocksymbols:
        dfsecond = pd.DataFrame()
        try:
            dfsecond = pd.read_json('https://api.iextrading.com/1.0/stock/'+symbolsecond+'/chart/5y')
        except Exception: pass #some symbols don't have json urls
        try:
            dailyreturnavgsecond = '{:f}'.format((dfsecond.changePercent-(1.76/365)).mean())
            standdevsecond = '{:f}'.format((dfsecond.changePercent-(1.76/365)).std())
        except AttributeError: pass #some symbol jsons don't have changePercent attributes
        try:
            covariance = ((((dfmain.changePercent)-((dfmain.changePercent-(1.76/365)).mean()))*((dfsecond.changePercent)-((dfsecond.changePercent-(1.76/365)).mean()))).mean())
            correlation = (((((dfmain.changePercent)-((dfmain.changePercent-(1.76/365)).mean()))*((dfsecond.changePercent)-((dfsecond.changePercent-(1.76/365)).mean()))).mean())/(((dfmain.changePercent-(1.76/365)).std())*((dfsecond.changePercent-(1.76/365)).std())))
            correlationlist.append(correlation)
            print(correlationlist)
        except AttributeError: pass #missing attribute
    dfcorrelation = pd.DataFrame(correlationlist)
    dfcorrelation.rename(index=str)
    dfbig = pd.concat([dfbig, dfcorrelation], axis=1, join='inner')


#create dfreturns and dfstanddev, concatenate them on dfbig
dfreturns = pd.DataFrame(returnslist)
dfstanddev = pd.DataFrame(standdevlist)
dfbig2 = pd.concat([dfbig, dfreturns], axis=1, join='inner')
dfbig2 = pd.concat([dfbig2, dfstanddev], axis=1, join='inner')

#correlation matrix columns so they line up
spacelist=[""]
headers = spacelist + stocksymbols
morenameslist=["Returns", "Standard Deviation"]
headers = headers + morenameslist
dfbig2.columns = headers
dfbig2.Returns = dfbig2.Returns.astype(float)
print(dfbig2)#For doing the rest in Excel, dfbig2.to_excel


weightlist = []
for stock in stocksymbols:
    weight = (1/len(stocksymbols))
    weightlist.append(weight)

weightdict = {'Ticker': stocksymbols}
dfweight = pd.DataFrame(weightdict)
dfweight['Weight'] = weightlist

portfolioreturns = (dfbig2["Returns"] * dfweight["Weight"])
print("Portfolio Return: ", sum(portfolioreturns))

dfbig = dfbig.drop(columns='Ticker')

weightlistlist = []
for x in weightlist:
    weightlistlist.append([x])
weightarray = (np.array(weightlistlist).T)
standdevarray = (np.array([standdevlist]).T)

weightedstd = np.dot(weightarray, standdevarray)

portfoliovariance = np.dot((weightarray),(np.dot(dfbig, weightarray.T)))
print("Portfolio variance: ", portfoliovariance)