When you work at a pod I often hear you have a dashboard tracking key industry data, like input costs for example when it comes to industrials or chemcials companies - and then a lot of this can feed into the models.
When you are actually building the models, what are the methods you use to zero in on whether your estimates are accurate or not? Do you actually try to get the raw numbers to tie in historically, or is more data analysis stuff to help with directionality?
Is it something like: Make estimates for what the make up of COGS is (essentially build up a custom index of COGS), and then run a regression to see how fitted the index change is with previous years, with some margin of error on this? So if this index declined 3% for 2Q, you would build in a 3% decline to COGS with a margin of error of +/- 0.5% or something?
Trending Content - Hedge Fund Forum
Career Resources
Bumperoni
Ya that’s about right. Have to understand the moving pieces to get to the why. It’s not going to be perfect but if you’re modeling out commodity chems for instance then you want to factor in your inputs at a granular level so that may mean running a regression, goal seeking, etc. Then it’s a matter chipping away at the uncertainty until you can accurately back-test and have high confidence you can nail normal quarters and then see how big your error is historically vs. outlier quarters. Outliers are tricky because they’re typically a lot of parallel factors so a one-off tie-out in your model could be dumb luck but if you’re directionally accurate (like within a std dev) on a consistent basis then that’s a win imo.
Back test, don’t usually move COGS directly. Usually use it in some type of margin bridge.
Could you elaborate?
Guessing they are saying don't tie it directly, see how fitted you get against past moves, and then use it for directionality/degree on margin bridges?
Suscipit tenetur illum deserunt est eius accusantium minus. Et nemo molestiae ullam. Quod libero dolores culpa ea est.
Et temporibus neque est eos. Recusandae ut laboriosam alias sunt.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...