Set out how multifactor models are applied and explain what the results of such analyses show

Multifactor models are used for return attribution, risk attribution, portfolio construction and strategic allocation. In return attribution, active return splits into a factor part, the sum of each active factor sensitivity times that factor's return, and a residual security-selection part. In risk attribution, active risk squared splits into active factor risk and active specific risk, and each factor's share shows where the bets lie. In construction, a manager can build a portfolio that matches a benchmark's factor sensitivities, or tilt deliberately towards one factor while neutralising the rest. The trap is to compute factor contributions using the portfolio's own sensitivities rather than its active sensitivities, which credits the benchmark's exposures to the manager.