Developing Financial Applications in F#

Learn to develop simple financial applications with F#, like the Twitter bot featured in this course.

F# is a functional-first programming language developed by Microsoft and used extensively in financial analysis and financial applications. F# expert developer Kit Eason steps you through the process of developing a simple F# financial application: a Twitter bot that charts stock price changes and respond to tweets with some simple descriptors of the stock performance, including gain/loss and highs/lows. Along the way, you'll learn the basics of F# syntax, including values, arrays, functions, and expressions, and how to test your code, analyze and chart third-party data. The lessons also provide a primer to concepts like test-driven development and railway-oriented programming—best practices for any F# development workflow.

Introduction

• Welcome
• Target audience and prerequisites
• Exercise files
• Working with software updates

1. Get Started with F#

• Defining values and calling functions
• Discriminated union and pattern matching
• Record type and arrays
• Forward piping
• Array mapping and iteration
• If-else expressions
• Exception handling
• Option types
• The bot and type signatures

2. Build a Simple Parser with Unit Testing

• Visual F# Power Tools and NCrunch
• Install NUnit
• Write a trivial NUnit test
• Write a real test
• Write a parser
• Refactor tests
• Handle unhappy paths with generic discriminated unions
• Test unhappy paths
• Extend and check test coverage
• Challenge
• Solution

3. Use F# CSV Type Provider to Get Data

• Recap and code tidy-up
• Introducing Quandl
• Using the CSV type provider to get data
• Wrapping the GetData function in a Choice.Result
• Challenge
• Solution

4. Analyze Data with F# Collection Functions

• Define a type to model a data summary
• Use collection functions to generate the summary
• Use .fsx scripts to run code experimentally
• Challenge
• Solution

5. Use RStats Provider and ggplot2 to Plot Data

• RStats and ggplot2
• Develop the chart generator
• Make a run harness for the chart generator
• Challenge
• Solution

6. Use BoxKite with Twitter

• Create a Twitter application
• Use credentials to connect to Twitter
• React to directed tweets

7. Deploy a Working Bot

• Assemble functions into a reply function
• Call the reply function
• Try the bot running locally
• Deploy the bot to Azure
• Understanding ROP

Conclusion

• Next steps