These 216 commits were written between 2014-10-04 and 2014-12-12.
These were my first attempts at building a financial market backtesting simulator from scratch.
There's a LOT of inefficient code that was never optimized, e.g., needless list traversals which scale really poorly as the number of trades in a session increase.
This was my first stab a backtesting simulator. I ran into a lot of design problems because I really had no idea of the scope of what I was trying to build. Go's type system didn't really help.
This was a port of the exchange_simulator to implement Walk Forward Optimization/Analysis as detailed by Robert Pardo in "The Evaluation and Optimization of Trading Strategies". I abandoned this in favor of redoing the simulator from scratch in Rust. See the "rust_simulator" repo.
These are a few random things I was testing when evaluating the idea of using Genetic Algorithm-based trading strategies.
These are a few utilities around training Neural Networks using libfann. I wanted to build some NN-backed indicators which trading strategies could use, but I never got that far with the simulator.
As of my last time touching this code (early Dec 2014), go build
is sufficient for building it. It doesn't use anything outside the stdlib except one of my own libraries on Github.
BSD