Update app.py
Browse files
app.py
CHANGED
@@ -1,76 +1,106 @@
|
|
1 |
import streamlit as st
|
2 |
-
import numpy as np
|
3 |
import pandas as pd
|
4 |
-
import
|
5 |
-
import
|
6 |
-
|
7 |
-
# Streamlit app layout
|
8 |
-
st.title("Estimating statistics from two lists of historical values")
|
9 |
-
|
10 |
-
with st.form("my_form"):
|
11 |
-
N = st.number_input("How many elements do you want for each list?", step=1)
|
12 |
-
|
13 |
-
# Button to generate the numbers
|
14 |
-
if st.form_submit_button("Click to generate random numbers' lists"):
|
15 |
-
# Generate two lists of 100 random numbers each
|
16 |
-
x = [random.randint(0, 20) for _ in range(N)]
|
17 |
-
y = [random.randint(0, 20) for _ in range(N)]
|
18 |
-
|
19 |
-
# Display the lists in the app
|
20 |
-
# st.write('List 1:', x)
|
21 |
-
# st.write('List 2:', y)
|
22 |
-
|
23 |
-
if "x" in globals():
|
24 |
-
x_bar = np.mean(x)
|
25 |
-
y_bar = np.mean(y)
|
26 |
-
|
27 |
-
st.subheader("Expected values")
|
28 |
-
st.write(f"E(x) = {x_bar:,.2f}")
|
29 |
-
st.write(f"E(y) = {y_bar:,.2f}")
|
30 |
-
|
31 |
-
var_x = np.var(x)
|
32 |
-
var_y = np.var(y)
|
33 |
-
|
34 |
-
st.subheader("Variances")
|
35 |
-
st.write(f"var(x) = {var_x:,.2f}")
|
36 |
-
st.write(f"var(y) = {var_y:,.2f}")
|
37 |
-
|
38 |
-
cov_xy = np.corrcoef(x, y)[0, 1]
|
39 |
-
|
40 |
-
st.subheader("Correlation")
|
41 |
-
st.write(f"corr(x, y) = {cov_xy:,.2f}")
|
42 |
-
|
43 |
-
plt.scatter(x, y)
|
44 |
-
plt.title("Correlations")
|
45 |
-
plt.xlabel("x")
|
46 |
-
plt.ylabel("y")
|
47 |
-
|
48 |
-
st.pyplot(plt)
|
49 |
-
|
50 |
-
plt.close()
|
51 |
-
|
52 |
-
plt.title("Plot of returns")
|
53 |
-
plt.plot(np.arange(len(x)), x)
|
54 |
-
plt.plot(np.arange(len(y)), y)
|
55 |
-
plt.ylim(-10, 40)
|
56 |
-
plt.xlabel('"time"')
|
57 |
-
plt.ylabel('"returns"')
|
58 |
|
59 |
-
|
60 |
|
61 |
-
|
62 |
-
var_x + var_y - 2 * cov_xy * np.sqrt(var_x * var_y)
|
63 |
-
)
|
64 |
-
w_y = 1 - w_x
|
65 |
|
66 |
-
|
67 |
|
68 |
-
|
69 |
-
st.write("Assuming x and y represent returns of portfolios")
|
70 |
-
st.write(f"w_x = {w_x:,.2f}")
|
71 |
-
st.write(f"w_y = {w_y:,.2f}")
|
72 |
-
st.write(f"E(r) = {w_x*x_bar + w_y*y_bar:,.2f}")
|
73 |
-
st.write(f"var(r) = {var_r:,.2f}")
|
74 |
|
|
|
|
|
|
|
75 |
else:
|
76 |
-
st.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import streamlit as st
|
|
|
2 |
import pandas as pd
|
3 |
+
import numpy as np
|
4 |
+
import optuna
|
5 |
+
import plotly.express as px
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
|
7 |
+
st.title("Portfolio weights calculator")
|
8 |
|
9 |
+
help_string = "NOTA: El formato utilizado aquí es llamando cada columna de GOOGLEFINANCE."
|
|
|
|
|
|
|
10 |
|
11 |
+
"Stocks - Sheet1.csv"
|
12 |
|
13 |
+
check_box = st.checkbox("¿Deseas usar el archivo precargado?")
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
+
if check_box:
|
16 |
+
uploaded_file = "Stocks - Sheet1.csv"
|
17 |
+
file_name = uploaded_file
|
18 |
else:
|
19 |
+
uploaded_file = st.file_uploader("Sube aquí tu archivo de excel", type=[".xls", ".xlsx", ".csv"], help=help_string)
|
20 |
+
file_name = uploaded_file.name
|
21 |
+
|
22 |
+
if uploaded_file is not None:
|
23 |
+
# Can be used wherever a "file-like" object is accepted:
|
24 |
+
if file_name[-3:] == "csv":
|
25 |
+
df = pd.read_csv(uploaded_file)
|
26 |
+
else:
|
27 |
+
df = pd.read_excel(uploaded_file)
|
28 |
+
|
29 |
+
df = df.drop(0, axis=0)
|
30 |
+
df = df.drop("Unnamed: 2", axis=1).drop("Unnamed: 4", axis=1).rename({"Unnamed: 0": "Date"}, axis=1)
|
31 |
+
|
32 |
+
df['Date'] = pd.to_datetime(df['Date']).dt.date
|
33 |
+
|
34 |
+
stocks = list(df.columns)[-3:]
|
35 |
+
stocks_rets = []
|
36 |
+
|
37 |
+
for i in stocks:
|
38 |
+
stocks_rets.append(i+"_ret")
|
39 |
+
df[i] = df[i].astype(float)
|
40 |
+
df[i+"_ret"] = (df[i] - df[i].shift(1)) / df[i].shift(1)
|
41 |
+
|
42 |
+
st.write(df[["Date"] + stocks_rets])
|
43 |
+
|
44 |
+
# Plotting with Plotly
|
45 |
+
fig = px.line(df, x=df.Date, y=stocks, labels={'value': 'Value', 'variable': 'Series'}, title='Time Series Plot')
|
46 |
+
fig.update_layout(xaxis_title='Date', yaxis_title='Value')
|
47 |
+
|
48 |
+
# Use Streamlit to render the plot
|
49 |
+
st.plotly_chart(fig)
|
50 |
+
|
51 |
+
ret_list = df[stocks_rets].mean().to_numpy().reshape(-1, 1)
|
52 |
+
cov_matrix = df[stocks_rets].cov().to_numpy()
|
53 |
+
|
54 |
+
optim_choice = st.selectbox("Elige la forma de optomizar :", ("max returns", "min variance", "max returns - variance"))
|
55 |
+
|
56 |
+
def portfolio_variance(weights, covariance_matrix):
|
57 |
+
return np.dot(weights.T, np.dot(covariance_matrix, weights))
|
58 |
+
|
59 |
+
def portfolio_returns(weights, expected_returns):
|
60 |
+
return np.dot(weights.T, expected_returns)
|
61 |
+
|
62 |
+
if optim_choice == "max returns":
|
63 |
+
def objective(trial):
|
64 |
+
w1 = trial.suggest_uniform('w1', 0, 1)
|
65 |
+
w2 = trial.suggest_uniform('w2', 0, 1)
|
66 |
+
w3 = 1 - w1 - w2
|
67 |
+
weights = np.array([w1, w2, w3]).reshape(-1, 1)
|
68 |
+
return np.dot(weights.T, ret_list)
|
69 |
+
|
70 |
+
study = optuna.create_study(direction="maximize")
|
71 |
+
study.optimize(objective, n_trials=100, show_progress_bar=True)
|
72 |
+
|
73 |
+
elif optim_choice == "min variance":
|
74 |
+
def objective(trial):
|
75 |
+
w1 = trial.suggest_uniform('w1', 0, 1)
|
76 |
+
w2 = trial.suggest_uniform('w2', 0, 1)
|
77 |
+
w3 = 1 - w1 - w2
|
78 |
+
weights = np.array([w1, w2, w3]).reshape(-1, 1)
|
79 |
+
return np.dot(weights.T, np.dot(cov_matrix, weights))
|
80 |
+
|
81 |
+
study = optuna.create_study(direction="minimize")
|
82 |
+
study.optimize(objective, n_trials=100, show_progress_bar=True)
|
83 |
+
|
84 |
+
else:
|
85 |
+
def objective(trial):
|
86 |
+
w1 = trial.suggest_uniform('w1', 0, 1)
|
87 |
+
w2 = trial.suggest_uniform('w2', 0, 1)
|
88 |
+
w3 = 1 - w1 - w2
|
89 |
+
weights = np.array([w1, w2, w3]).reshape(-1, 1)
|
90 |
+
return np.dot(weights.T, ret_list) - np.dot(weights.T, np.dot(cov_matrix, weights))
|
91 |
+
|
92 |
+
study = optuna.create_study(direction="maximize")
|
93 |
+
study.optimize(objective, n_trials=100, show_progress_bar=True)
|
94 |
+
|
95 |
+
w1 = study.best_params['w1']
|
96 |
+
w2 = study.best_params['w2']
|
97 |
+
w3 = 1- w1 - w2
|
98 |
+
|
99 |
+
weights = np.array([w1, w2, w3]).reshape(-1, 1)
|
100 |
+
|
101 |
+
yearly_returns = (1 + np.dot(weights.T, ret_list)[0, 0]) ** 252 - 1
|
102 |
+
yearly_variance = np.dot(weights.T, np.dot(cov_matrix, weights))[0, 0] * 252
|
103 |
+
|
104 |
+
st.write(f"Los pesos son: :green[{stocks[0]} -> {w1:,.4f}], :green[{stocks[1]} -> {w2:,.4f}], :green[{stocks[2]} -> {w3:,.4f}]")
|
105 |
+
st.write(f"El retorno anualizado del portafolio es: :green[{yearly_returns:,.4f}]")
|
106 |
+
st.write(f"La varianza anualizado del portafolio es: :green[{yearly_variance:,.4f}]")
|