This time series course is the first in a four-course curriculum that covers mean reversion- and momentum-based trading strategies before concluding with a focus on machine learning and algorithmic trading.

By the end of this course, students will be able to:

• Manipulate time series data sets using Python's Pandas library
• Identify the components of a time series
• Use tools such as moving average, exponentially-weighted moving average, and the Hodrick-Prescott filter to identify long-term trends
• Perform auto-correlation on time series data
• Determine stationarity of a time series
• Create an ARMA model to forecast financial data
• Create an ARIMA model to forecast financial data
• Use ACF and PACF plots to estimate the order of ARMA and ARIMA models
• Model and predict volatility with GARCH
• Create a time series linear regression model using Scikit-learn
• Analyze and predict seasonal effects using regression
• Quantify the accuracy of a linear regression model
• Define overfitting and articulate the benefits of parsimony
• Use Scikit-learn to train and test time series data
• Make out-of-sample predictions in rolling windows
• Evaluate the accuracy of out-of-sample predictions