Udacity AI for trading course

Personal repository for storing quizes and notebooks from Udacity AI for trading nano degree.

2. Folder description

Lesson 2

Stock prices

Lesson about stocks and common terminology. stock_data.upynb contains example of stock data set: open, close, high and low prices.

Lesson 3

Market mechanics

resample_data.ipynb shows how to resample time data, e.g. get close price every 3rd day or every week.

Lesson 5

Stock returns

Contains a quiz asking to compute returns for each ticker and date in close.

Lesson 6

Momentum trading

dtype.ipynb contains an example how to generate position (Long 30 share of stock when the price is above 50 dollars or hort 10 shares when it's below 20 dollars). top_and_bottom_performing.ipynb shows how to filter out stocks by their performance.

Lesson 10

Regression

test_normality.ipynb shows how to check whatever distribution is normal using Shapiro-Wilk or Kolmogorov-Smirnov methods. regression.ipynb investigates to stocks using regression.

Lesson 11

Time series modeling

Example of autoregressive moving average (ARIMA) fit on time series data is shown in autoregression_quiz.ipynb.

Lesson 12

Volatility

rolling_windows.ipynbshows how to annualize volatility using rolling window method.

Lesson 13

Pairs trading and mean reversion

pairs_cadidates.ipynb shows how to check if two stocks prices correlate in time.

Lesson 15

Stocks, Indices, Funds

Shows how to calculate cumulative returns.

Lesson 17

Portfolio risk and return

Shows how to calculate covariance matrix.

Lesson 18

Portfolio optimization

Shows to to solve portfolio optimization problems using cvxpy package.

Lesson 22

Factors

Introduces to open source Zipline package which is used to create pipelines for stock algorithmic trading.

Lesson 24

Risk factor models

Notebooks provides answers to quizzes how to calculate covariance of assets, factor model returns on assets or portfolio, and variance of portfolio or historical data.

Lesson 26

Risk factor models with PCA

Introduces to principle component analysis (PCA) usage on factors.

Lesson 27

Alpha factors

Shows how to smooth alpha factors, calculate factor quantiles and ranks, how to calculate sharpie ration, transfer coefficient, z-scores.

Lesson 28

Alpha factor research methods

Provide examples,how to implement alpha factors into trading strategy using overnight returns or regression against time as alpha factors.

Project 1

Trading with Momentum Project

A very simple trading strategy is implemented on quotemedia historical stock data project_1_starter.ipynb. The main idea is to buy top 10 best performing and short top 10 worst performing stocks.

Project 2

Breakout strategy

In project_2_starter.ipynb breakout strategy is implemented, which buys or shorts stocks, which prices beat 5, 10 or 20 day period historical prices. Signal filtering is also implemented to reduce number of trades.

Project 3

Smart Beta and portfolio optimization

In this project, betas (risk factors) are calculated for stock portfolio based on paid dividends amd returns. Then portfolio optimization is done to minimize variations. Finally, this portfolio is compared to benchmark index.

Project 4

Alpha research and factor modeling

In this project, statistical risk model is built using PCA. Then model is expanded using 5 alpha factors:

Momentum 1 Year Factor the hypothesis "Higher past 12-month (252 days) returns are proportional to future return.",
Mean Reversion 5 Day Sector Neutral Factor, the hypothesis "Short-term outperformers(underperformers) compared to their sector will revert."
Mean Reversion 5 Day Sector Neutral Smoothed Factor just smoothed factor,
Overnight Sentiment Factor, hypothesis "overnight returns reveals information about investors sentiments",
Overnight Sentiment Smoothed Factor just smoothed factor.

These factors are then combined into Zipline pipeline. Model is then evaluated using factor-weighted returns, quantile analysis, sharpe ratio, and turnover analysis. Finally, the stock portfolio is optimized using multiple optimization formulations.

Requirements

All requirements are available on requirements.txt file.

TK-Problem/nano_degree_AT