Access the forward looking adjustments page for any pool by clicking on the forward looking adjustment figure on the main page. Access the custom qualitative adjustments by clicking on that column's number in the appropriate pool.
Total qualitative factor adjustments are shown at the bottom right of the forward-looking adjustment page, and they are a combination of the forward-looking adjustments and custom qualitative factor adjustments listed directly above.
Custom qualitative factor adjustments are determined by the bank. The components of the custom qualitative factor adjustment can be seen by clicking on the underlined custom factors number near the bottom right of the page or the black custom factors button at the top right of the page. The custom factors are pool-specific, and the total reserve for the pool and the bank are updated immediately in the model when factors are added or updated. Click the home button on the navigation bar at the top of the page to see the impact of the custom qualitative factors. These factors are carried forward for subsequent reports, so they do not need to be entered again, just updated for the new period.
The forward-looking adjustment calculations are listed in the amortization schedule at the top of this page. The same columns as the pool summary pages are included, but with an additional column for projected losses. Projected losses are determined for the first several years and then revert back to the same loss rate as historical losses.
The basis of the projected losses for the first several years can be viewed by clicking on the underlined projected losses than are different that the historical losses. All of the underlined projected loss links go to the same page.
At the top left of this page, the black line in the graph shows an objective economic indicator that has the best stastical fit with historical losses for this type of loan for the group of peer banks used to calculate average historical losses. Potential leading indicators at the national and state level are listed below and used to determine the highest correlation. The graph is flipped upside-down, so good numbers are on the bottom and bad numbers are on the top (the scale for the leading indicator line is on the right side of the graph). Historical losses for the peer group for each quarter are plotted with the orange line.
The black line helps to project what will eventually happen with the orange line. At the bottom left of the page, a new graph shows the same lines, but the black leading indicator line is shifted to the right by the time period that provides the best statistical fit, somewhere between one and three years. This graph provides an overview of how the two lines match up when accounting for the lag period between the leading indicator and the eventual losses for the peer banks for this type of loan.
At the top right of the page is a scatter-plot of the leading indicator (x-axis) and historical losses after the lag period (y-axis). Each dot represents an individual historical quarter. The blue line is a linear regression line that represents the best fit for the historical data. It minimizes the distance between each of the dots and the line. This regression line is what is used to project future losses.
At the bottom right of the page is some statistical data about the regression line. The coefficient of determination (also know as r-squared) shows the statistical correlation of the regression line with the historical data. A correlation of 1.0 would be a perfect correlation, and the closer the number is to 1, the better the statistical fit. The projected loss amount is calculated by taking the most recent leading indicator figure from the graph at the top left of the page, and finding that number along the x-axis of the graph at the top right of the page. If a straight vertical line is traced from that point on the x-axis to the regression line and then a horizontal line is drawn to the y-axis, it will cross at the projected loss figure. A minimum projected loss of 0.10% is used for all pools.
The projected loss on the graph page is used for the lag period that provides the best statistical fit in the amortization schedule on the qualitative factor page. If the lag period is one year, the first year of projections will show the projected loss rate. Year two losses are the average of the projected loss rate and the historical loss rate, and year three projected losses revert back to the historical loss rate. If the lag period is two years, the second year of the projection schedule is the projected loss rate and the first and third years are the average of the historical loss rate and the projected loss rate. If it is a 1 1/2 year lag period, years one and two are 3/4 of the projected loss rate and 1/4 of the historical loss rate. Projected losses are totalled and compared to total historical losses, and the difference is the forward-looking qualitative factor adjustment (sometimes the difference between the total figures are off slightly due to rounding).
Possible leading indicators include (we consider both the 4 quarter rolling average and the current ratio for these indicators):