Python for Financial Analysis and Algorithmic Trading

Install and set up the python for finance environment with Anaconda, then activate the py finance conda environment and run a Jupyter notebook.

Refresh your Python skills quickly with a concise crash course designed to get you moving through the Python for financial analysis and algorithmic trading material.

The Python crash course offers a quick refresher for new or returning programmers, covering syntax essentials and guidance on using Jupyter Notebook or any IDE, with exercise and solutions notebooks.

Master Python fundamentals for financial analysis, including numbers, strings, lists, dictionaries, tuples, sets, booleans, comparison and logical operators, control flow, loops, functions, and print formatting in a Jupyter notebook.

Explore Python basics in this crash course part two, mastering comparison and logical operators, if-elif-else control flow, for and while loops, the range function, and an introduction to list comprehension.

Master Python crash course basics by building functions, using lambda expressions, and applying map and filter to sequences. Learn about parameters, return values, docstrings, and essential string, list, dictionary methods.

Explore the Python crash course exercises in a Jupyter notebook, practicing square root of price, indexing and slicing, string formatting, nested dictionary access, and creating simple data-processing functions.

Learn practical Python basics through hands-on exercises, such as computing square roots, slicing strings, and formatting output. Practice navigating nested dictionaries and building functions to detect, count, and average prices.

Explore NumPy, the numerical Python library at the core of the Python for finance ecosystem, and learn the basics that underpin other related libraries.

Learn NumPy, a numerical Python library for fast data handling with arrays, mastering indexing and operations to prepare you for Pandas in quantitative analysis.

Import numpy as np to work with arrays, including vectors and matrices, and create zeros, ones, linspace, identity, and random samples for two-dimensional financial data.

Explore NumPy operations through element-wise array arithmetic, broadcasting, and universal functions in a Jupyter notebook, including sqrt, exp, max, sin, and log.

Learn numpy indexing and selection with bracket and slice notation, explore broadcasting and view versus copy semantics, and practice two-dimensional array indexing and conditional selection for data analysis.

Explore NumPy basics with arrays and matrices through practical exercises on creation, indexing, operations, random number generation, and computing sums, standard deviation, and axis-based column sums.

Develop numpy proficiency by creating zeros, ones, fives, and ranges; generate arange, reshape into matrices, practice indexing, slicing, and axis-based sums, and use linspace and seeds for reproducibility.

Explore pandas, the powerful panel data library for Python, and see how it simplifies dealing with data. You'll discover why it quickly becomes one of your favorite Python libraries.

Explore pandas fundamentals with series and dataframes, read and write data from CSV, Excel, SQL, and more, and perform fast, group-by operations for financial datasets.

Explore pandas series, with named indices and flexible data types, by creating them from lists, arrays, or dictionaries and indexing by label or position, paving the path to dataframes.

Explore Pandas data frames built from series with a shared index, then select, create, and drop columns and rows using loc and iloc for precise indexing.

Master conditional selection in pandas with boolean masks to filter rows or columns, use ampersand and pipe for multiple conditions, and reset or set the index.

Explore pandas multi-index dataframes: build a two-level index from tuples, name index levels, index from outside in with .loc, and use cross section for multi-level selection.

Explore missing data handling in pandas by creating a dataframe, dropping rows or columns with NaN using dropna and axis, applying thresh, and filling gaps with fillna or column means.

Group by in pandas aggregates rows by a chosen column and applies numeric functions such as mean, sum, and std, returning a per-group numeric result while ignoring non-numeric columns.

Explore how to merge, join, and concatenate data frames in pandas with pd.concat and pd.merge, including axis selection, inner joins, and key-based merging for multi-source datasets.

Explore essential pandas operations for financial analysis, including finding unique values with unique and n unique, counting with value_counts, conditional selection, apply with lambdas, and pivot table creation.

Learn how to input and output data with pandas data frames, focusing on csv files and commands like read_csv and to_csv, plus optional excel, html, and sql workflows.

Practice pandas skills with the failed banks dataset: import pandas, read_csv into a DataFrame named banks, view head and columns, and list states via one-line tasks.

Explore pandas exercise solutions by inspecting dataframes, retrieving columns and head, computing unique values, grouping and sorting, and using lambdas to analyze bank data.

Learn to visualize financial data using numpy, pandas, and matplotlib with pandas built-in visualization tools; create custom plots in a few lines of code to quickly gain insights.

Learn visualization in Python with matplotlib and pandas visualizations built on top of it, focusing on time series plots, static visuals, and a Jupyter trick to interact with plots.

Explore Matplotlib, a Python 2d plotting library for publication quality figures. Learn the functional and object oriented APIs, basic plotting with plt, subplots, and axes in Jupyter and other environments.

Explore matplotlib's object-oriented plotting with plt.subplots for multi-panel figures, control figure size and dpi, add titles and labels, save figures, and create informative legends.

Explore how to customize Matplotlib plots with colors, line widths and styles, markers, and alpha transparency, and control axes with set_xlim and set_ylim for precise visualizations.

Practice Matplotlib through a guided set of exercises, building figures, axes and subplots with X, Y and Z data, while reviewing inline plotting in a Jupyter notebook and previewing solutions.

Master object-oriented matplotlib by creating figures, adding axes, and plotting x, y, and z with labeled axes, titles, and styles, including subplots and figure sizing.

Explore pandas visualization capabilities to generate histograms, area, bar, line, scatter, hexbin, and density plots from data frames with the time series index, plus optional seaborn styling and matplotlib customization.

Learn to visualize time series with pandas and matplotlib, plot adjusted close and volume, control axes with x and y limits, and create interactive plots in Jupyter notebooks.

Recreate the plots from the notebook using df3 as CSV source and review the pandas visualization exercises. Practice inserting cells below to preserve figures, with solutions in the next lecture.

Practice Pandas visualization in a Jupyter notebook by creating scatter plots, histograms, box plots, kde, and area charts, and styling with ggplot, sizing, and legends.

Learn two ways to fetch financial data with Python on your computer: pandas data reader and Quandl. Understand api access, data frames, and csv fallbacks for regional or firewall issues.

Use the Pandas data reader in a Jupyter notebook to fetch stock data from Google Finance and other APIs, building a data frame with open, high, low, close, and volume.

Explore Quandl's free core financial data, navigate the wiki end-of-day prices, and use the Python API in a Jupyter notebook to fetch, filter, and export stock and real estate data.

Explore essential pandas tools for working with time series financial data and build skills to analyze market trends using Python for financial analysis and algorithmic trading.

Explore pandas time series with date time indices, resampling, time shifts, and rolling and expanding methods to analyze stock price data over time.

Learn to use a date time index in pandas for time series data by converting Python date time objects into a date time index, and inspect day and month attributes.

Explore time resampling with pandas to convert date strings to datetime, set a date-time index, and aggregate by frequency—end of year, quarterly, or custom first-day analytics.

Shift data with pandas' shift method to align price data across time steps; illustrate forward and backward shifts, handling nulls and lost tail, with a date time index.

Learn how to apply pandas rolling and expanding to financial time series, create moving averages with windows like 7, 30, and visualize with plots, plus Bollinger bands.

Complete a full financial analysis capstone project to review the libraries covered and learn daily returns, using an exercise notebook with tasks and questions and solution videos for code along.

Learn to analyze stock market data in capstone project using Python, NumPy, pandas, and matplotlib, comparing returns and cumulative daily returns for Tesla, GM, and Ford from 2012 to 2017.

Import stock data for Tesla, Ford, and GM using pandas data reader or read CSV. Visualize open prices and volume, compute total traded, moving averages, and a scatter matrix.

Create a candlestick chart by combining a documentation example with provided data, and learn to prepare date axes and OHLC tuples for a Ford January 2012 dataset.

Explains daily returns calculations for Tesla, Ford, and GM, using shift and pct_change, then analyzes volatility with histograms, density estimation plots, box plots, and scatter matrices to explore stock correlations.

Explore cumulative returns and daily returns calculations in a stock market analysis project, using Pandas and Jupyter Notebook to compare time-series investments.

Explore the unique properties of time series data, learn key components, and study time series algorithms such as ARIMA, while considering why some models may not suit financial securities.

Explore time series fundamentals with Pandas and Statsmodels, learn about ETS, EMA, and ARIMA models, and understand when these techniques suit financial data and when they do not.

Define time series concepts, including trends, stationary behavior, seasonality, and cycles, with examples from Google trends and stock data, and introduce statsmodels for analysis.

Learn how to use the Statsmodels library for time series analysis in Python, including accessing datasets, descriptive tests, plotting, and applying the Hodrick-Prescott filter to extract trend and cycle.

Explore ETS theory by decomposing time series into error, trend, and seasonality with Statsmodels, visualizing observed values, trend, seasonality, and residuals from the airline passengers dataset.

Discover how exponentially weighted moving averages (EWMA) improve on simple moving averages by weighting recent values in pandas to reduce lag and clarify trends.

Explore how to use exponentially weighted moving averages with pandas to reduce lag and highlight recent trends, plotting the ewma alongside simple moving averages in a time series notebook.

Decompose a time series with ETS in Python using statsmodels to separate trend, seasonality, and residuals, compare additive and multiplicative models, and visualize results in a Jupyter notebook.

Arima models, including non-seasonal and seasonal Arima, and learn how differencing and the augmented Dickey-Fuller test assess stationarity for forecasting time series with autoregressive, integrated, and moving average components.

Explore autocorrelation functions (acf) and partial autocorrelation functions (pacf) to identify ar and ma components, determine p, d, q terms, and guide arima modeling in python.

Explore arima and seasonal arima using statsmodels on a real time-series dataset of monthly milk production. Visualize, make the series stationary, examine acf/pacf, and practice forecasting.

Use the augmented Dickey-Fuller test to assess stationarity, interpret p-values, and implement a reusable ADF check in Python with statsmodels, then apply first, second, and seasonal differencing.

Learn to generate and interpret autocorrelation and partial autocorrelation plots for multiple time series differences using statsmodels and pandas, then prepare for seasonal arima forecasting in the next lecture.

Implement seasonal ARIMA in statsmodels, fit the model, review residuals, and forecast future values, noting ARIMA's limits for financial data and Brownian motion.