As part of a scenario testing exercise, I want to test the P&L impact of a 100 bps upward parallel shift in real rates on a portfolio of public equities. This will be calculated in two separate scenarios, where:
1) rates rise due to high growth causing inflation
2) rates rise during a period of high inflation and low growth (stagflation)
So the three factors are GDP growth, inflation and real interest rates, and each scenario has a set of specific values that each factor will take on (e.g in scenario 1, real rates +1%, growth +1.75%, inflation +2% from current value). I need to measure my portfolio's sensitivity to these macroeconomic factors, or to the increase in rates while controlling for the regime implied by the other two factors.
What are some practical ways that I can do this? I can regress historic equity returns for my equities on each factor to get an indication of their sensitivity, but it is generally a poor fit. Are there examples in the literature of similar exercises, or an approach that is typically used?