Add new figures and comparisons to VAR charts

[matdehaven]

• [x] Plot frequency domain figures with the period on the x-axis, instead of frequency
• [x] Create a figure that shows FEV just around the business frequencies
• [x] Create a variance-weighted average (compare that to a cumulative sum of the FEV over cummulative sum total)
• [x] Check if we are rescaling for the FEV figure in Time Domain. The y-axis seems small for output.

[matdehaven]

• [x] Check how BCA calculated their FEV averages

[matdehaven]

BCA paper calculates the "Variance Contributions" at business cycle frequencies as the ratio between the sums of the variance explained over total variance at business cycle frequencies. This is the same as the variance-weighted average of the forecast error variance decomposition.

[matdehaven]

The Forecast error variance calculations are correct. The sum of the frequency FEV is equal to the limit of the time domain FEV. These should both be equal to the unconditional variance of each variable in the VAR, but they are not for this VAR. I think it is because some of the variables are in levels and are thus not stationary.