List of Skills Taught:

SAS

Interface

• Introduction to the SAS interface

Basics

• Comments in SAS
• Permanent & temporary folders
• Introduction to data steps
• Introduction to procedures
• Viewing and printing data
• Basic functions & operators
• If-then statements
• Creating indicator variables
• Keep & drop statements
• Where statements
• Missing data
• Date basics
• Time intervals
• Sorting data
• Duplicate observations
• Creating lag and lead variables
• Creating count variables
• Creating sub-samples
• Stacking data
• Merging with a data step
• Merging with Proc SQL
• Merging and dates
• Summarizing data using Proc Means
• Summarizing data using Proc SQL
• Calculating buy-and-hold raw returns
• Ranking variables
• Creating groups or portfolios
• Regrouping across time
• Summarizing data with groups or portfolios

Outliers

• Winsorizing & trimming
• Robust regression

MACRO Basics

• Creating MACRO variables
• Creating and running MACROS
• MACRO repositories

WRDS

• Remote connecting to WRDS
• Running code at WRDS
• Pulling Compustat data
• Pulling CRSP data
• Merging Compustat and CRSP
• Uploading & downloading data

Common Statistics Procedures

• Proc Means, Univaritate, Corr Ttest and Npar1way
• Proc reg (along with fixed effects, clustered errors, predicted and residual values)
• Proc surveyreg, logistic surveylogistic, genmod, qlim, and robustreg

Python

Basics

1. Setting up Python
2. Anacondas installation
3. Common libraries (pandas, numpy)
4. Basic commands
5. Basic markdown
6. Reading in, importing, exporting data
7. Data types, dataframes and data configuration
8. Merging data, stacking data
9. Subsetting, dropping data
10. Cleaning data, missing values, backfilling, winsorize/truncate
11. Exploring data, descriptive statistics
12. Variables: creating variables, dropping variables, leads and lags, dealing with dates, ranked variables
13. Grouping, frequency tables
14. Basic regression

WRDS

1. Installing the WRDS library
2. Connecting to WRDS via Python
3. Employing SQL within Python
4. Acquiring Compustat data
5. Creating common financial statement variables
6. Acquiring CRSP data
7. Calculate cumulative and buy-and-hold abnormal returns
8. Merging Compustat and CRSP

Web scraping

1. Introduction to HTML structure
2. Identifying HTML xpaths
3. Scraping with scrapy
4. Splitting HTML as text
5. Pandas read_html
6. Interacting with a web browser using Selenium (sending text to search boxes, clicking buttons, extracting page source)
7. Scraping EDGAR
8. EDGAR file management and master IDX files
9. Introduction to EDGAR XBRL data
10. Scraping Yahoo! Finance

Robotic process automation

1. Obtaining and validating user input
2. Automating the keyboard (basic keyboard functions, special keys)
3. Automating the mouse (position, movement, clicking, scrolling)
4. Basics of operating system file management
5. Deleting files
6. Copying files
7. Moving files
8. Listing files in a directory
9. Creating folders

Textual analysis:

1. Introduction to natural language processing
2. Text functions (replace, strip, upper, lower, count, join, find, startswith, endswith)
3. Introduction to regular expressions (match, search, split, findall, sub, groups)
4. Tokenization (word_tokenize, sent_tokenize, regexp_tokenize)
5. Text pre-processing (stemming, stop words, punctuation)
6. Counting words
7. Calculating disclosure tone
8. Calculating the Fog Index

Advanced Python:

1. Machine learning
2. Intro to machine learning
3. What is machine learning?
4. Types of machine learning models: supervised vs. unsupervised
5. Applications of machine learning in practice
6. Applications of machine learning in academic research
7. The basic machine learning process
8. Logistic regression
9. Introduce machine learning concepts for predicting binary dependent variables.
10. Predict default using financial statement and stock price data
11. True negatives, true positives, false negatives, false positives
12. Accuracy vs. sensitivity vs. specificity
13. statsmodels vs. scikit-learn in Python
14. Exercise: predict employee attrition
15. Linear Regression
16. Introduce machine learning concepts for continuous dependent variables
17. Predict CDS spreads using financial statement and stock price data
18. Evaluating the prediction
19. Exercise: predict future revenue growth
20. Random forest classification
21. Introduction to decision trees
22. From decision trees to random forest (including a discussion of overfitting)
23. RandomForestClassifier from scikit-learn
24. Feature importance
25. Exercise: predict employee attrition
26. Random forest regression
27. Difference between random forest classification and random forest regression
28. RandomForestRegressor from scikit-learn
29. Exercise: predict future revenue growth
30. Neural network classification
31. What are neural networks (e.g., input layer, hidden layers, output layer, etc.)
32. MLPClassifier from scikit-learn
33. Standardizing input variables
34. Permutation feature importance
35. Exercise: predict banking customer churn
36. Neural network regression
37. Difference between neural network classification and neural network regression
38. MLPRegressor from scikit-learn
39. Exercise: Predict housing prices
40. Support vector classification
41. The basics of support vector machines (SVM) (e.g., hyperplanes, data separation, margin, kernels, decision boundaries, cost parameter, support vectors)
42. SVC from scikit-learn
43. Standardizing input variables
44. Exercise: predict banking customer churn
45. Support vector regression
46. Difference between support vector classification and support vector regression
47. SVR from scikit-learn
48. Cherkassky and Ma (2004) method for obtaining epsilon and C parameters
49. Exercise: Predict housing prices
50. Cross validation and Grid Search
51. What is cross validation?
52. Holdout cross validation
53. K-fold cross validation
54. Data independence in K-fold cross validation
55. Rolling window cross validation for panel data
56. Hyperparameter tuning with grid search
57. Exercise: predict default using cross validation techniques
58. Machine learning and textual analysis introduction
59. Machine learning with a large number of independent variables (e.g., words or ngrams)
60. CountVectorizer from sklearn
61. Stop words
62. Stemming using the Porter Stemmer algorithm
63. Creating matrices for use in the machine learning models
64. Feature importance
65. Exercise: predict restaurant review ratings based on comments
66. Machine learning and textual analysis – TF-IDF
67. An introduction to term-frequent inverse document frequency (tf-idf)
68. TfidfVectorizer from scikit-learn
69. Exercise: predict restaurant review ratings based on comments using tf-idf
70. Machine learning and textual analysis – Creating customized vectorizers
71. Limitations with built-in vectorizers
72. How to create your own customizable vectorizer
73. Sparse data
74. Running textual analysis machine learning models on documents stored in your file system
75. Exercise: predict restaurant review ratings based on a customized vectorizer
76. Latent Dirichlet Allocation
77. Unsupervised machine learning model
78. What is Latent Dirichlet Allocation (LDA) and how does it work?
79. genism
80. Exercise: Implement LDA on a set of earnings conference call transcripts
81. Machine learning project
82. Use the machine learning models taught in the course to predict daily abnormal stock returns using conference call transcripts
83. ChatGPT
84. Introduction to ChatGPT
85. What is ChatGPT and why you should use it in coding
86. An overview with examples of how to use ChatGPT to enhance your coding (e.g., to write, optimize, and debug code)
87. Use ChatGPT to write code
88. Step by step guide to writing prompts and interacting with responses to write code
89. Example: We ask ChatGPT to write code to extract Apple’s net income from its 2022 10-K on the SEC’s EDGAR site.
90. Exercise: Use ChatGPT to write code to extract the tone of words spoken by analysts on earnings conference calls.
91. Use ChatGPT to optimize code
92. Multiple examples of inefficient code with prompts to ChatGPT to optimize and remove inefficiencies.
93. Exercise: Write code to answer basic coding questions. Then use ChatGPT to optimize any inefficiencies in your code
94. Use ChatGPT to debug code
95. Multiple examples of coding errors with prompts to ChatGPT to debug and remove these errors.
96. Integrated Development Environments
97. Introduction to Integrated Development Environments (IDEs)
98. Python Idle
99. Examples of IDEs for Python
100. Reasons for using a more sophisticated IDE
101. Visual Studio Code
102. Overview of the Visual Studio Code environment
103. Explorer Window
104. Extensions
105. Color Themes
106. Command Palette
107. Commands and features (e.g., intellisense, multiple cursors, clicking functions, warnings, errors, etc.)
108. Debugging
109. Jupyter Notebook Extension
110. Markdown vs. code cells
111. Debugging within cells
112. Keyboard shortcuts and video tutorials
113. PyCharm
114. PyCharm Installation
115. Overview of the PyCharm Environment
116. Projects explorer
117. Debugging
118. Find Action
119. Color Themes
120. Commands and features (e.g., intellisense, multiple cursors, clicking functions, wrapping code, extract method, etc.)
121. Local history to restore deleted code

STATA

Basics

1. Understanding the STATA environment
2. Reading files from SAS and Python
3. Viewing a dataset
4. Generating variables
5. Replacing variables
6. Using local variables
7. Keeping and dropping variables
8. Sorting data
9. Using for loops

Model Output

1. Running basic regression models
2. Creating Excel output tables using outreg2
3. Modifying Excel output table options
4. Adding statistics to Excel output tables (Adj. R2, Pseudo R2)
5. Appending additional columns
6. Adding column headers
7. Excluding variables from Excel output tables