Tools for financial simulation

Efficient quasi-event-driven backtesting functions with opinionated input requirements.

Designed according to the following philosophy:

• Prioritise backtesting efficiency
• Function inputs should be close to the natural outputs of the quant research process: dataframes/matrixes of prices and target weights.
• Include helper functions for efficiently wrangling these research outputs into formats suitable for fast backtesting

The easiest way to install and load rsims is using pacman::p_load_current_gh which wraps devtools::install_github and require:

pacman::p_load_current_gh("Robot-Wealth/rsims", dependencies = TRUE)

Depends on application:

• min_commission_backtest: suitable for backtesting under the assumption of minimum commission, for instance many equities brokers.
• fixed_commission_backtest: suitable for backtesting under the assumption of linear (fixed percentage) commission, for instance many crypto exchanges.
• fixed_commission_futs_backtest: suitable for backtesting expiring products under the assumption of linear (fixed percentage) commission, for instance CME futures.

These functions are optimised for efficiency. They simulate trading into a set of weights (calculated upstream) subject to transaction cost and other constraints. They expect matrixes for prices and target weights.

Example usage:

library(rsims)
results <- min_commission_backtest(prices, theo_weights, trade_buffer = 0., initial_cash = 10000, commission_pct = 0, capitalise_profits = FALSE)

Further details in the respective function documentation.

Examples of wrangling data for input to these functions can be found in the vignettes.

Position deltas are calculated using the trade buffer approach. Positions are rebalanced once they deviate from their target by more than a user-supplied trade_buffer. Rebalancing happens slightly differently depending on the commission model used:

• for minimum commission backtesting, rebalance back to the target weight
• for fixed commission backtesting, rebalance back to the target weight plus/minus the trade buffer

These are heuristic rules that are theoretically optimal give the different cost models. Here's a good derivation from @macrocephalopod on Twitter.

Other approaches to trade determination may be implemented in the future.

Currently rsims implements simplified cost models that only include commission. For some applications, the various costs (market impact, spread, and commission) might be reasonably represented by such a model. No attempt is made (yet) to explicitly account for these costs separately. Borrow and funding costs are not yet implemented.

TODO