This repository contains my solution to the Sigma Internship Coding Challenge, designed to demonstrate proficiency with the Quantrocket platform, coding abilities, and understanding of basic trading algorithms.

Installation: It is instructed to install Quantrocket either locally or on a cloud platform.

Data Retrieval: Pulling the daily close price data for AAPL for the year 2023, leveraging Quantrocket's freely available US stock price data.

The Trading Model
Objective: The core objective is to build a model that decides whether to place a buy order trade for the next day (d+1), aiming to maximize the portfolio's value based on daily returns (r(d)).

Calculating daily returns: r(d) = (p(d) - p(d-1))/p(d-1).

Classifying states based on returns:
Bull (+1) if r(d) >= 0.01.
Flat (0) if -0.01 < r(d) < 0.01.
Bear (-1) if r(d) <= -0.01.

Applying a value function where the portfolio value (V(d+1)) is adjusted based on the state transitions from day d to d+1.
Increase the portfolio value by 1 if Flat on (d) to Bull state on (d+1)
Decrease the portfolio value by 1 if Flat on (d) to Bear state on (d+1) 
Stays unchanged for all other scenarios

This repository includes a Jupyter Notebook (Final.ipynb) containing the complete code implementation of the trading model described above. The notebook walks through the data retrieval process, calculation of daily returns, state classification, application of the value function, and decision-making to identify optimal buy points.

The final portfolio value V(N) after executing the trading strategy throughout 2023.
A list of optimal buy indices, indicating the days on which buy orders were placed.

Portfolio Value =  17
Buy Indices:  [6, 8, 12, 16, 21, 28, 30, 41, 50, 52, 59, 61, 69, 79, 85, 88, 94, 100, 103, 108, 110, 113, 117, 120, 123, 133, 142, 160, 164, 177, 187, 191, 207, 209, 212, 216, 218, 232, 234, 238]

if previous_state == 'Flat':    

    prob_flat_to_bull = transition_probabilities['Flat']['Bull'] # Check the probability of transitioning from Flat to Bull
    
    if prob_flat_to_bull > 0.5: # Example criterion for making a buy decision, and this threshold is arbitrary and should be optimized
    
    print("High probability of transitioning from Flat to Bull. Consider placing a buy order.")

pivot = (high + low + close) / 3
bottom_cpr = (high + low) / 2
top_cpr = (pivot - bottom_cpr) + pivot
momentum_of_stock = "High" if abs(bottom_cpr - top_cpr) < threshold else "Low" 

# Bullish Crossover & Narrow CPR
if macd[0] > macd[1]  and macd[2] > macd[3] and momentum_of_stock == "High":
    direction = "Strong Uptrend"
# Bullish Crossover & Wide CPR
elif macd[0] > macd[1]  and macd[2] < macd[3] and momentum_of_stock == "Low":
    direction = "Weak Uptrend"
# Bearish Crossover & Narrow CPR
elif macd[0] < macd[1]  and macd[2] > macd[3] and momentum_of_stock == "High":
    direction = "Strong Downtrend"
# Bearish Crossover & Wide CPR
elif macd[0] < macd[1]  and macd[2] < macd[3] and momentum_of_stock == "Low":
    direction = "Weak Downtrend"