Overview of Programming Languages for Quantitative Finance/Trading

First of all, let's state one simple thing: programming language is just a tool. If you can build the same spreadsheet with Excel and LibreOffice, it means you can solve the same task with different programming languages too. In theory, it means you can learn any modern programming language and be capable of solving any task with it.

In practice, some programming languages provide better or simpler solutions for your problems. And to compare the languages, we need to state these problems clearly:

1. Simplicity - It's simpler to write code in a high-level programming language than in a low-level one. By "simpler," I mean it has a low entry threshold if you're a beginner and is less time-consuming if you're already an advanced programmer.
2. Performance - If you need to implement complex algorithms or work with big data, performance becomes crucial. But in reality, performance usually depends more on implementation than on the programming language itself.
3. Ecosystem - Useful frameworks and libraries bring algorithms and solutions so you don’t have to reinvent the wheel.
4. Popularity - The more popular the chosen programming language, the more articles you can find on the internet, the better community support, the better ecosystem, etc. It works like a snowball effect.

JavaScript

JavaScript is a great choice for beginners and is undeniably popular (top 1). It's the best proof of the importance of popularity: it has tons of frameworks for data analysis, machine learning, mathematical calculations, etc., despite not being designed for those due to technical limitations.

However, it can be a strong tool for data visualization, with libraries like D3.js and Highcharts providing interactive charts right in the browser.

Short answer - Narrow application.

P.S. TypeScript doesn’t make any significant change in this regard.

Python

Python is one of the most popular programming languages with a low entry threshold, much like JavaScript. But there are two facets that make Python much better suited for any data-related activity:

1. Python has the most powerful data science and machine learning ecosystem, outclassing JavaScript and any other programming language by a wide margin. Python is a must-have language for data processing.
2. Although Python is relatively slow compared to low-level programming languages, many popular libraries (like pandas and numpy) are compiled in C, allowing us to leverage the performance of C within Python.

These points make Python strong in all four facets we discussed. It makes Python the most recommended general-purpose programming language in this field.

R, Scala, Julia

Other programming languages popular for data processing can be good additions to your tech stack. However, they're not as widely recognized or popular as Python.

If you have a strong reason to use one of these languages, go for it. They're still common and suitable and can be a better choice if you really understand why you need them.

C++, Rust

Highly efficient and performant, C++ is suitable for high-frequency trading (HFT) systems. It provides fine-grained control over system resources and memory but at the cost of complexity and more time-consuming implementation compared to other languages.

This makes C++ the best programming language for HFT and great for solving specific tasks requiring strong performance in general.

Rust has seen great growth in recent years but still lacks the strong ecosystem and community support for financial analysis that C++ has. It can be considered a rising star.

Java, C#, Go, and Other Languages

These are common backend languages that are sometimes chosen for big production applications. They can be considered intermediate options between the slower Python and the hardcore C++.

Conclusions

Python is a strong favorite for any data processing tasks, with quantitative analysis being just a part of that. Choosing it as your first programming language is never a mistake.

Other programming languages can be useful for solving specific tasks more effectively (with better performance or frameworks), but you need to understand why you need them.

Choosing a second language strongly depends on the kind of problems you're trying to solve.