Struct TurnoverOptimizer 

pub struct TurnoverOptimizer;

Finds the minimal set of trades that moves a portfolio from current weights to target weights while keeping total execution cost low.

Algorithm

1. For each symbol in the union of current and target weights, compute the weight delta Δw = target - current.
2. If |Δw| < tolerance, skip (already within band).
3. Otherwise add a Trade with the estimated cost for a unit-notional order sized proportionally to |Δw|.
4. Trades are sorted by |Δw| descending (largest rebalance first).

Implementations

impl TurnoverOptimizer

pub fn optimize( current: &HashMap<String, f64>, target: &HashMap<String, f64>, cost_params: &CostParams, tolerance: f64, ) -> Vec<Trade>

Generate the minimal set of trades to rebalance from current to target.

tolerance is the minimum absolute weight change that warrants a trade (e.g. 0.005 = 50 bps). Smaller changes are ignored to avoid excessive round-trip cost.

cost_params are used to estimate the cost of each trade assuming a unit notional of $1, with shares = |Δw| / price where price is set to 1.0 for weight-space estimation.