+ Reply to Thread
Results 1 to 2 of 2

Thread: Double Regression?

  1. #1
    Points: 3,358, Level: 36
    Level completed: 6%, Points required for next Level: 142

    Thanked 0 Times in 0 Posts

    Double Regression?

    I am looking at two years of financial info by month. I need to project out next month - however, at the end of our one fiscal year there was a significant negative financial entry. by leaving this as is, it has a large impact on the regression.
    since this large entry is really due to an on going accounting error, does it make statistical sense to spread out this entry over the affected time period and apply regression on these new variables?
    if so, what is the logical way of doing this?
    Thank you very much

  2. #2
    TS Contributor
    Points: 14,811, Level: 78
    Level completed: 91%, Points required for next Level: 39
    Miner's Avatar
    Greater Milwaukee area
    Thanked 405 Times in 363 Posts

    Re: Double Regression?

    Since this is a definite outlier with a know cause, it is appropriate to correct it. How to correct it is the question. If it is a systemic error, determine whether it is a flat amount or a percentage of some base value then allocate the error appropriately across the time periods. This is the best option because deleting the value would bias your intercept at a minimum and probably your slope as well.

  3. The Following User Says Thank You to Miner For This Useful Post:

    bugman (11-21-2014)

+ Reply to Thread


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts

Advertise on Talk Stats