app.py

import streamlit as st
import numpy as np
import matplotlib.pyplot as plt
import random

def main():
    st.title("Statistics for two historical lists🏦")

    with st.form("my_form"):
        N = st.number_input("Insert the total values for the lists:", step=1)

        # Button to generate the numbers
        if st.form_submit_button("Click to generate random numbers' lists"):
            # Generate two lists of random numbers each
            lista1 = [random.randint(0, 20) for _ in range(N)]
            lista2 = [random.randint(0, 20) for _ in range(N)]

            # Button to perform analysis
            if len(lista1) != 0 and len(lista2) != 0:
                analizar_inversiones(lista1, lista2)
            else:
                st.write(":red[Please, generate random numbers' lists]")

def analizar_inversiones(lista1, lista2):
    # Convertir las listas en arrays numpy
    array1 = np.array(lista1)
    array2 = np.array(lista2)

    # Calcular desviación estándar
    std_dev1 = np.std(array1)
    std_dev2 = np.std(array2)

    # Calcular valor esperado
    mean1 = np.mean(array1)
    mean2 = np.mean(array2)

    # Calcular correlación
    correlation = np.corrcoef(array1, array2)[0, 1]

    # Crear gráfica
    fig, ax = plt.subplots()
    ax.scatter(array1, array2)
    ax.set_xlabel('Lista 1')
    ax.set_ylabel('Lista 2')
    ax.set_title('Investment graph')
    ax.grid(True)
    st.pyplot(fig)

    # Decidir a cuál invertir más y a cuál menos
    if mean1 > mean2:
        invertir_mas = "List 1"
        invertir_menos = "List 2"
    else:
        invertir_mas = "List 2"
        invertir_menos = "List 1"

    # Mostrar resultados con un formato agradable
    st.write("Results of the Analysis:")
    st.write(f"Standard deviation of list 1: {std_dev1}")
    st.write(f"Standard deviation of list 2: {std_dev2}")
    st.write(f"Estimated value list 1: {mean1}")
    st.write(f"Estimated value list 2: {mean2}")
    st.write(f"Correlation: {correlation}")
    st.write(f"Invest more in: {invertir_mas}")
    st.write(f"Invest less in: {invertir_menos}")

    # Calculate weights and portfolio returns
    x_bar = mean1
    y_bar = mean2
    var_x = std_dev1**2
    var_y = std_dev2**2
    cov_xy = np.cov(array1, array2)[0, 1]

    w_x = (x_bar - y_bar + var_y - cov_xy * np.sqrt(var_x * var_y)) / (
        var_x + var_y - 2 * cov_xy * np.sqrt(var_x * var_y)
    )
    w_y = 1 - w_x

    var_r = (w_x**2)*var_x + (w_y**2)*var_y + 2*w_x*w_y*cov_xy*np.sqrt(var_x * var_y)

    st.subheader("Weights and portfolio returns")
    st.write("Assuming x and y represent returns of portfolios")
    st.write(f"w_x = {w_x:,.2f}")
    st.write(f"w_y = {w_y:,.2f}")
    st.write(f"E(r) = {w_x*x_bar + w_y*y_bar:,.2f}")
    st.write(f"var(r) = {var_r:,.2f}")

if __name__ == "__main__":
    main()