Mr Kissell and Mr Glantz use timing risk to represent the uncertainty of the transaction cost estimate.
The two main sources of uncertainty are volatility of both the asset’s price and of the trade volume but there are also other factor like spread risk especially if we are in fast market like in the last months where with Mr Donald Trump we have a lot of changes.
With this formula we can resume it with a generic error factor E
Timing risk = volatility risk + liquidity risk + E
Price volatility is the most important risk. The more volatile an asset then the more likely its price will move away and so increase the transaction costs.
The liquidity risk represents the uncertainty with respect to the market impact cost. Market impact cost are estimated based on historical volumes, so if the actual trading volumes differ significantly this may result in a shift in the market impact.
If the market volume is higher then the market impact costs will tend to be less. Based on the calculation for timing cost shown before , timing risk is effectively the residual cost once the price trend has been accounted for, in mathematical formula
Timing risk = ∑ ( xj * ( mj – p*j))
Where xj is the size of each execution, mj is the mid price at the time and p*j is the expected price based on the price trend.
Timing riks is a focus for the risk based algorithms, most execution tactics don’t directly incorporate timingrisk, some become aggressive execution as time goes on, to try to complete any extant orders.
An aggressive VWAP algorithm will still schedule its orders based on a historic volume profile, so orders will still be spaced out over time.