A couple of things that we want to bring to your attention before you start creating and running your FDIC models in order to make your models successful: 1) You want to complete your regression model for the Year Ended 12/31/20. 2) All variables MUST be divided by the annual Average Total Assets (ATA) of each bank. This is to take into account the size difference between each bank. You will need an annual ATA for each bank, meaning that you will need to download 12/31/19 total asset data in addition to 12/31/20 so that you can calculate an average amount for the year. Use Excel to make sure you are using V-Lookups and other formulas to make this easy for yourself!!! 3) For each X variable you have the choice of using the level amount (the actual amount at 12/31/20) or a change in the account (variance between 12/31/19 and 12/31/20). You decide which one makes more sense for your model. If you decide to use the change for the year then you will obviously need to download 12/31/19 data as well so you can calculate the change for the year. 4) Your very first X variable needs to be 1/ATA. This is a size variable so that you can compare banks the size of JP Morgan to the small local bank down the street. Otherwise, your regression will be all skewed. So, your final regression should look like this: (Total Accruals/ATA) = (1/ATA)+(X1/ATA)+(X2/ATA)+(X3/ATA)…