-
Notifications
You must be signed in to change notification settings - Fork 0
/
main.py
209 lines (175 loc) · 9.03 KB
/
main.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
import streamlit as st
import numpy as np
import pandas as pd
import requests
import matplotlib.pyplot as plt
import os
import datetime
from prophet import Prophet
import requests
import pandas as pd
st.set_page_config(
page_title="Capstone Project - Investment and Trading",
page_icon="🧊",
initial_sidebar_state="expanded",
menu_items={
'Get Help': 'https://www.linkedin.com/in/paulo-mota-955218a2/',
'Report a bug': "https://www.linkedin.com/in/paulo-mota-955218a2/",
'About': "Compare and predict US stocks!"
}
)
def get_stock_data_by_symbol(symbol: str):
url = "https://alpha.financeapi.net/symbol/get-chart?period=MAX&symbol=" + symbol
token = os.getenv('YAHOO_TOKEN')
payload = {}
headers = {
'X-API-KEY': token
}
response = requests.request("GET", url, headers=headers, data=payload)
response_json = response.json()
if 'attributes' in response_json:
return pd.DataFrame(response_json['attributes']).T
else:
return pd.DataFrame()
def fill_missing_values(dataframe):
return dataframe.ffill().bfill()
def get_data(symbols, dates):
df_final = pd.DataFrame(index=dates)
for symbol in symbols:
df_temp = get_stock_data_by_symbol(symbol)
if not df_temp.empty:
df_temp = df_temp[['close']]
df_temp.index = pd.to_datetime(df_temp.index)
df_temp = df_temp.rename(columns={"close": symbol})
df_final = df_final.join(df_temp)
return df_final
def get_data_for_training(symbol, start_date, end_date):
"""Read stock data (adjusted close) for given symbols from CSV files."""
df_final = pd.DataFrame(index=pd.date_range(start_date, end_date))
# file_path = symbol_to_path(symbol)
df_temp = get_stock_data_by_symbol(symbol)
if df_temp.empty:
raise 'Symbol invalid or with empty stock data!'
df_temp = df_temp[['close']]
df_temp.index = pd.to_datetime(df_temp.index)
df_final = df_final.join(df_temp)
df_final = df_final.ffill().bfill()
return df_final
def compute_daily_returns(df):
"""Compute and return the daily return values."""
# TODO: Your code here
# Note: Returned DataFrame must have the same number of rows
return df[:-1] / df[1:].values - 1
def get_data_for_training(symbol, start_date, end_date):
"""Read stock data (adjusted close) for given symbols from CSV files."""
df_final = pd.DataFrame(index=pd.date_range(start_date, end_date))
df_temp = get_stock_data_by_symbol(symbol)
if not df_temp.empty:
df_temp = df_temp[['close']]
df_temp.index = pd.to_datetime(df_temp.index)
df_final = df_final.join(df_temp)
df_final = df_final.ffill().bfill()
return df_final
return False
st.title("Capstone Project Data Scientist: Investment and Trading")
st.markdown(
"Autor: Paulo Mota [ LinkedIn](https://www.linkedin.com/in/paulo-mota-955218a2/)")
st.markdown("")
st.info(
"This software utilizes stock symbols from the USA and retrieves financial data from Yahoo Finance. More info: https://financeapi.net/")
tab_compare, tab_predict = st.tabs(["Compare Stocks", "Predictor"])
with tab_compare:
st.header("Closing price and daily returns comparator")
today = datetime.date.today()
tomorrow = today + datetime.timedelta(days=1)
yesterday = today - datetime.timedelta(days=1)
start_default = yesterday - datetime.timedelta(days=30)
start_date = st.date_input('Start date', datetime.date(1993, 1, 1), min_value=datetime.date(1993, 1, 1))
end_date = st.date_input('End date', yesterday)
if not start_date < end_date:
st.error('Error: End date must fall after start date.')
if end_date > yesterday:
st.error('Error: End date must be equal or older than yesterday.')
# get symbols from user
symbol_list = st.text_input("Enter Symbols Separated by comma (,)", value='GOOG,TSLA,MSFT',
key='symbols').upper().split(',')
st.info('The chart will not display invalid symbols or symbols that are not listed in the USA market.', icon="⚠️")
if st.button('Run'):
valid_input = True
if not start_date < end_date:
st.error('Error: End date must fall after start date.')
valid_input = False
if end_date > yesterday:
st.error('Error: End date must be equal or older than yesterday.')
valid_input = False
if symbol_list[0] == '':
st.error('Error: You must have at least one stock symbol.')
valid_input = False
with st.spinner("Processing..."):
# check if all fields are filled
if valid_input:
dates = pd.date_range(start_date, end_date) # date range as index
df_data = get_data(symbol_list, dates) # get data for each symbol
# Fill missing values
df_data = fill_missing_values(df_data)
st.header('Closing price by stock symbol')
st.line_chart(df_data)
with st.expander("What is closing price?"):
st.write(
"""The closing price is the raw price or cash value of the last transacted price in a security before the market officially closes for normal trading. It is often the reference point used by investors to compare a stock's performance since the previous day""")
daily_returns = compute_daily_returns(df_data)
st.header('Daily returns by stock symbol')
st.line_chart(daily_returns)
with st.expander("What is daily return?"):
st.write(
"""Daily return is calculated by subtracting the opening price from the closing price. If you are calculating for a per-share gain, you simply multiply the result by your share amount. If you are calculating for percentages, you divide by the opening price, then multiply by 100.""")
else:
st.error("Please fill all fields!")
with tab_predict:
st.header("Closing price predictor with Prophet model")
st.write("More info about Prophet: https://facebook.github.io/prophet/")
st.write("Why should you predict stock value?")
st.write(
"""Stock value prediction is to forecast the future price or value of a stock, which is influenced by various factors such as company performance, market conditions, and global economic trends. By accurately predicting the future value of a stock, investors can make informed decisions about whether to buy, sell, or hold a particular stock, potentially resulting in significant financial gains.""")
stock_symbol = st.text_input('Stock Symbol', 'TSLA')
if st.button('Predict Stock'):
with st.spinner("Processing..."):
today = datetime.date.today()
start_date = today - datetime.timedelta(days=1095)
end_date = today - datetime.timedelta(days=1)
predict_days = 365
future_date = end_date + datetime.timedelta(days=365)
df_train = get_data_for_training(stock_symbol, start_date, end_date)
print(df_train)
if not df_train is False:
# Preparing training data
df_train = df_train[['close']]
df_train = df_train.reset_index()
df_train = df_train.rename(columns={'index': 'ds', 'close': 'y'})
df_train.tail()
# Create Prophet model
model = Prophet()
# Train model
model.fit(df_train)
# Make predictions on test data
future = model.make_future_dataframe(periods=predict_days, freq='D')
predictions = model.predict(future)
predictions = predictions.rename(columns={'ds': 'date'})
predictions = predictions.rename(columns={'yhat': 'prediction'})
predictions = predictions[['date', 'prediction']]
predictions = predictions.set_index('date')
df_train = df_train.rename(columns={'ds': 'date'})
df_train = df_train.rename(columns={'y': 'close_value'})
df_train = df_train[['date', 'close_value']]
df_train = df_train.set_index('date')
df_final = pd.DataFrame(index=pd.date_range(start_date, future_date))
df_final = df_final.join(predictions)
df_final = df_final.join(df_train)
st.header('Closing price forecast - ' + stock_symbol)
st.line_chart(df_final)
with st.expander("How I calculated that?"):
st.write(
""" I am utilizing the Prophet model to analyze the past three years of historical data, beginning from yesterday, with the objective of forecasting the closing price for a duration of one year.""")
else:
st.error('Error: The stock symbol provided either does not exist or is not listed in the US market.',
icon="🚨")