Add files via upload

project3_yutingji.py

@@ -0,0 +1,254 @@
+# Project Option 3. Long-and-Short Investment in S&P 500 Stocks
+# Benjamen Graham Value Stock Criteria include:
+# 1) Credit rating of B+ and above
+# 2) Lowest price to book ratio
+# Of course, Goldman Sachs can also sell stocks with low credit ratings and high market-to-book at the same time
+
+# User Interface:
+# 1. The user can enter the day range (such as 2020-01-01 to 2021-09-01)
+# 2. The user can also choose the industry from the stocks in S&P 500 index
+# Within the chosen industry:
+# 3. Your code finds a list of stocks satisfying the Benjamen Graham Value Stock Criteria (Modified Altman Z-Score above A- level)
+#    and another list of stocks having a Modified Z-Score below BBB- level, both in year 2020
+# 4. Your code also finds a list of stocks with market-to-book (M2B) above 20 and another list with M2B below 2.0
+# 5. Among these four lists of stocks (high- vs low-credit rating and low- vs. high-M2B), your code will long $50,000 in a stock
+#    that is both in the high-credit list and in the low-M2B list on the first day of your chosen day range
+#    and short $50,000 in a stock that is both in the low-credit list and in the high-M2B list 
+# 6. Find one pair of stocks with the highest investment return in dollars during the chosen day range
+
+# Modified Z-Score
+# M-Z-score = 3.25 + 6.56A + 3.26B + 6.72C + 1.05D
+# A = (Current Assets - Current Liabilities) / Total Assets
+# B = Retained Earnings / Total Assets
+# C = EBIT / Total Assets
+# D = Book Equity / Total Liabilities
+
+# Rating 	Minimum Z-score
+# AAA		8.80
+# AA		8.40
+# A+		8.22
+# A		6.94
+# A-		6.12
+# BBB+		5.80
+# BBB		5.75
+# BBB-		5.70
+# BB+		5.65
+# BB		5.52
+# BB-		5.07
+# B+		4.81
+# B		4.03
+# B-		3.74
+# CCC+		2.84
+# CCC		2.57
+# CCC-		1.72
+# D		0.05
+
+
+import random as rd
+import pandas as pd
+import numpy as np
+import yfinance as yf
+import datetime as dt
+import tkinter as tk
+import matplotlib.figure as fig
+import matplotlib.backends.backend_tkagg as bac
+
+def DownloadData(ticker, begin_day, end_day):
+    try:
+        dt.datetime.strptime(begin_day, '%Y-%m-%d')
+        dt.datetime.strptime(end_day, '%Y-%m-%d')
+        mydata = yf.download(ticker, begin_day, end_day)
+        return mydata['Adj Close']
+    except:
+        print('Date must be entered as YYYY-MM-DD')
+        return 0
+
+# Load all U.S. stocks in all sectors:
+stock_list = pd.read_csv('sp500.csv')
+
+def main():
+    # calculate stocks' z-score and market to book ratio
+    # and categories them
+    industry=myindustry.get()
+    begin_day=bday.get()
+    end_day=eday.get()
+
+    high_rating={}
+    low_rating={}
+    high_M2B={}
+    low_M2B={}
+    # establish industry data frame
+    new_frame = stock_list[stock_list['Industry']==industry]
+    tickers = new_frame[['Ticker','Company']].reset_index(drop=True)
+
+    count=0
+    print("Loading......")
+    for i in tickers['Ticker']:
+        count+=1
+        # make sure less than 30 runs to avoid being identified as a hacker
+        if count<30:
+            #find z-score by calling function zscore()
+            #find market to book by calling function M2B()
+            z=zscore(i)
+            m=M2B(i)
+            print(i,z,m)
+            if z>=6.12:
+                high_rating[i]=z
+            elif z<=5.07 and z!=-1:
+                low_rating[i]=z
+            if m>=20:
+                high_M2B[i]=m
+            elif m<=2 and m!=0:
+                low_M2B[i]=m
+    print('High Rating:',high_rating)
+    print('Low Rating:',low_rating)
+    print('High M2B:',high_M2B)
+    print('Low M2B:',low_M2B)
+
+    longdict={}
+    shortdict={}
+    # find qualifying stocks to long
+    for i in high_rating:
+        if i in low_M2B:
+            stockdata=DownloadData(i,begin_day,end_day)
+            longdict[i]=(stockdata[-1]-stockdata[0])*(round(50000/stockdata[0]))
+    # find qualifying stocks to short
+    for i in low_rating:
+        if i in high_M2B:
+            stockdata=DownloadData(i,begin_day,end_day)
+            shortdict[i]=(stockdata[0]-stockdata[-1])*(round(50000/stockdata[0]))
+    print("Longdict:",longdict)
+    print("Shortdict:",shortdict)
+
+    # append best pair into best_pair list
+    best_pair=[]
+    if len(longdict)!=0 and len(shortdict)!=0:
+        # find the key corresponding to the max value within that dictionary
+        lkey=max(longdict,key=longdict.get)
+        skey=max(shortdict,key=shortdict.get)
+        best_pair.append(lkey)
+        best_pair.append(skey)
+
+    # display result both in "best pair" box
+    # and in the scroll box
+    SetSelect(best_pair,tickers)
+
+def M2B(my_ticker):
+    # calculate market to book value based on 2020 data
+    begin_day=bday.get()
+    end_day=eday.get()
+    fin_data = yf.Ticker(my_ticker)
+    so = fin_data.get_info()['sharesOutstanding']
+    price_dataframe = fin_data.history(period='1y')['Close']
+
+    # get Market Value
+    mv = so*price_dataframe.loc['2020-12-31']
+    # get Book Value
+    balance_sheet = fin_data.get_balancesheet()
+    # if 2020 book value is found, calculate market to book ratio
+    if '2020' in balance_sheet:
+        current_asset = balance_sheet.loc['Total Current Assets']['2020']
+        current_liab = balance_sheet.loc['Total Current Liabilities']['2020']
+        bv=current_asset[0]-current_liab[0]
+        M2B = mv/bv
+    else:
+        # if 2020 book value is not found, skip the stock
+        M2B = 0
+
+    return M2B
+
+def zscore(my_ticker):
+    # calculate z-score based on 2020 data
+    fin_data = yf.Ticker(my_ticker)
+    income_stmt = fin_data.get_financials()
+    balance_sheet = fin_data.get_balancesheet()
+
+    if '2020' in balance_sheet and '2020' in income_stmt and 'Retained Earnings' in balance_sheet.index:
+    # Ratio of Working Capital / Total Assets (WC_TA)  #A
+        current_asset = balance_sheet.loc['Total Current Assets']['2020']
+        current_liab = balance_sheet.loc['Total Current Liabilities']['2020']
+        total_asset = balance_sheet.loc['Total Assets']['2020']
+        wc_ta = (current_asset[0] - current_liab[0]) / total_asset[0]
+
+    # Ratio of Retained Earnings / Total Asset (RE_TA) #B
+        retain_earning = balance_sheet.loc['Retained Earnings']['2020']
+        re_ta = retain_earning[0] / total_asset[0]
+
+    # Ratio of EBIT / Total Asset (EBIT_TA)  #C
+        ebit = income_stmt.loc['Ebit']['2020']
+        ebit_ta = ebit[0] / total_asset[0]
+
+    # Ratio of Equity MV / Total Liabilities (MV_TL)   #D
+        total_liab = balance_sheet.loc['Total Liab']['2020']
+        so = fin_data.get_info()['sharesOutstanding']
+        price_dataframe = fin_data.history(period='1y', interval='1d')['Close']
+        mv_tl = so * price_dataframe.loc['2020-12-31'] / total_liab
+
+        # if has data for 2020, calculate the Z-Score
+        z_score = 3.25 + 6.65*wc_ta + 3.26*re_ta + 6.72*ebit_ta + 1.05*mv_tl
+        return z_score[0]
+    else:
+        # if doesn't have data for 2020, return -1, skipping the ticker
+        return -1
+
+
+## Functions for TK interface
+def SetSelect(best_pair,tickers):
+    # if the best pair dictionary is not empty, show result
+    # else, show "no..." msg in Entry
+    if len(best_pair)!=0:
+        # show best pair in Entry
+        result='Long:'+best_pair[0]+'; Short:'+best_pair[1]
+        mypair.set(result)
+        # show best pair and its full company name in Scroll Box
+        myselect.delete(0,tk.END)
+        for s in best_pair:
+            comp = tickers.loc[tickers.Ticker==s,'Company'].values[0]
+            myselect.insert (tk.END, s+', '+comp)
+    else:
+        mypair.set('No best pair found')
+def ClearInput():
+    bday.set('')
+    eday.set('')
+    myindustry.set('')
+def QuitNow():
+    mywindow.quit()
+    mywindow.destroy()
+
+
+mywindow = tk.Tk()
+mywindow.geometry('600x600')
+mywindow.title('Porject 3')
+
+myframe = tk.Frame(mywindow)
+myframe.pack(side = tk.RIGHT)
+myscroll = tk.Scrollbar(myframe, orient=tk.VERTICAL)
+myselect = tk.Listbox(myframe, yscrollcommand=myscroll.set)
+myselect.pack(side=tk.LEFT)
+myscroll.config(command=myselect.yview)
+myscroll.pack(side=tk.RIGHT, fill=tk.Y)
+
+tk.Label(mywindow, text = 'Industry').place(x=30, y=10)
+myindustry = tk.StringVar()
+tk.Entry(mywindow, textvariable = myindustry).place(x=100, y=10)
+
+tk.Label(mywindow, text = 'Start Date').place(x=30, y=50)
+bday = tk.StringVar()
+tk.Entry(mywindow, textvariable = bday).place(x=100, y=50)
+
+tk.Label(mywindow, text = 'End Date').place(x=300, y=50)
+eday = tk.StringVar()
+tk.Entry(mywindow, textvariable = eday).place(x=370, y=50)
+
+tk.Label(mywindow, text = 'Best Pair').place(x=150, y=90)
+mypair = tk.StringVar()
+tk.Entry(mywindow, textvariable = mypair).place(x=220, y=90)
+
+tk.Button(mywindow,text="RUN", command = main).place(x= 60, y=190)
+tk.Button(mywindow,text="CLEAR", command = ClearInput).place(x=60, y=230)
+tk.Button(master=mywindow, text="Quit Now", activeforeground='red',command= QuitNow).pack(side=tk.BOTTOM)
+
+mywindow.mainloop()
+
+
+