(This might involve running a baseline regression for the two sets of knowledge independently, then generating comparisons of baseline-predicted consumption with true consumption for the two sets of data independently, then combining the outcomes alongside one another at the top. This information do