Spring 2019 Q2 (a)

Comments

• Graham

  March 2021

  First, I should point out that they definitely intended for you to use 1-step trending. You know this because they only provided information about historical trends. There was no further information given that would enable you to estimate prospective trends. And in fact the 2-step trending used the same 4% trend for both historical and prospective periods. If the trends are the same then there is no reason to use 2-step trending.

  But getting back to your question:

  • The problem was not stated clearly. It isn't clear whether the given data for AWP @ CRL is for quarters (as it's labelled) or for half-years. I think it's probably meant to be half-years.
  • That means the latest available data point is the second half of 2018, not 2018Q4, and the midpoint of that half-year would be Oct 1, 2018. But they also seem to be arguing that this data point may not be the appropriate "trend-to" date given that the "trend-from" date is Apr 1, 2017, which is in the first half of 2017. It looks like they don't want to trend from the first half of 2017 to the second of 2018. Instead, the "trend-to" date is the latest available similar data point, which would be the first half of 2018. And the midpoint of that period is Apr 1, 2018.

  But again, the intent was to use 1-step trending so all of these complications with 2-step trending could have been avoided. I think the reason a lot of people used 2-step trending is the way the data was given. Often, when you're given AWP @ CRL, they do want you to use 2-step trending. In this problem however, there was no way to estimate a prospective trend that was different from the historical trend so 2-step trending has no advantage. If you compare the final answer for sample answers 1 & 2, you'll see they are very close.