Update app.py

app.py

@@ -1,38 +1,79 @@
 import streamlit as st
 import numpy as np
+import pandas as pd
 import matplotlib.pyplot as plt
+import random
 
 # Streamlit app layout
-st.title("Random Sample Generator and Statistics Calculator")
-
-with st.form("my_form"):
-    sample_size = st.number_input("Enter the sample size:", min_value=1, step=1, value=100)
-    low = st.number_input("Enter the lower bound of the uniform distribution:", value=0)
-    high = st.number_input("Enter the upper bound of the uniform distribution:", value=1)
-    if st.form_submit_button("Generate Samples"):
-        # Generate random samples
-        samples = np.random.uniform(low, high, sample_size)
-
-    # Display the generated samples
-    if "samples" in locals():
-        st.subheader("Generated Samples")
-        st.write(samples)
-
-        # Calculate and display descriptive statistics
-        st.subheader("Descriptive Statistics")
-        st.write(f"Mean: {np.mean(samples):.2f}")
-        st.write(f"Standard Deviation: {np.std(samples):.2f}")
-        st.write(f"Variance: {np.var(samples):.2f}")
-        st.write(f"Min: {np.min(samples):.2f}")
-        st.write(f"Max: {np.max(samples):.2f}")
-
-        # Visualize the distribution with a histogram
-        st.subheader("Histogram")
-        plt.hist(samples, bins=20, color='skyblue', edgecolor='black')
-        plt.xlabel('Value')
-        plt.ylabel('Frequency')
-        plt.title('Histogram of Random Samples')
-        st.pyplot(plt)
-        plt.close()
-    else:
-        st.warning("Please generate samples first.")
+st.title("Estimating statistics from two sets of historical data")
+
+with st.form("data_input_form"):
+    N = st.number_input("Enter the number of data points for each set:", step=1)
+
+    # Button to generate random data sets
+    if st.form_submit_button("Generate Random Data Sets"):
+        # Generate two sets of N random data points each
+        set1 = [random.randint(0, 20) for _ in range(N)]
+        set2 = [random.randint(0, 20) for _ in range(N)]
+
+if "set1" in globals():
+    # Calculating mean of each set
+    set1_mean = np.mean(set1)
+    set2_mean = np.mean(set2)
+
+    st.subheader("Expected Values")
+    st.write(f"Mean of Set 1: {set1_mean:,.2f}")
+    st.write(f"Mean of Set 2: {set2_mean:,.2f}")
+
+    # Calculating variances
+    set1_var = np.var(set1)
+    set2_var = np.var(set2)
+
+    st.subheader("Variances")
+    st.write(f"Variance of Set 1: {set1_var:,.2f}")
+    st.write(f"Variance of Set 2: {set2_var:,.2f}")
+
+    # Calculating correlation coefficient
+    corr_coefficient = np.corrcoef(set1, set2)[0, 1]
+
+    st.subheader("Correlation Coefficient")
+    st.write(f"Correlation Coefficient: {corr_coefficient:,.2f}")
+
+    # Scatter plot
+    plt.scatter(set1, set2)
+    plt.title("Scatter Plot")
+    plt.xlabel("Set 1")
+    plt.ylabel("Set 2")
+
+    st.pyplot(plt)
+    plt.close()
+
+    # Line plot
+    plt.title("Line Plot of Data Sets")
+    plt.plot(np.arange(len(set1)), set1, label="Set 1")
+    plt.plot(np.arange(len(set2)), set2, label="Set 2")
+    plt.ylim(-10, 40)
+    plt.xlabel("Time")
+    plt.ylabel("Returns")
+    plt.legend()
+
+    st.pyplot(plt)
+    plt.close()
+
+    # Calculating weights and portfolio returns
+    w_set1 = (set1_mean - set2_mean + set2_var - corr_coefficient * np.sqrt(set1_var * set2_var)) / (
+        set1_var + set2_var - 2 * corr_coefficient * np.sqrt(set1_var * set2_var)
+    )
+    w_set2 = 1 - w_set1
+
+    portfolio_var = (w_set1**2)*set1_var + (w_set2**2)*set2_var + 2*w_set1*w_set2*corr_coefficient*np.sqrt(set1_var * set2_var)
+
+    st.subheader("Weights and Portfolio Returns")
+    st.write("Assuming Set 1 and Set 2 represent returns of portfolios")
+    st.write(f"Weight of Set 1: {w_set1:,.2f}")
+    st.write(f"Weight of Set 2: {w_set2:,.2f}")
+    st.write(f"Expected Portfolio Return: {w_set1*set1_mean + w_set2*set2_mean:,.2f}")
+    st.write(f"Portfolio Variance: {portfolio_var:,.2f}")
+
+else:
+    st.write("Please input the number of data points for each set.")