Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Comparison of Volume between two time periods using what test?

    Hi, I am trying to see if there's a statistical test to compare volumes for an item from two different time periods.

    For example, looking at differences between avg. monthly volume for the entire 2018 to 2019 period compared to monthly volume for each month after the start of an intervention (e.g. COVID-19). I have monthly volume for a specific item and it ranges from 2018 to 2020.

    I was told that I could do:

    Code:
    by id, sort : ttest volume, by(post) unequal welch
    However, I'm concerned that:
    1. This may not be the right approach (e.g. might require quasi-experimental approach)
    2. Degrees of freedom becomes a concern when comparing 1:1 or 1:n


    For simple descriptive statistics purposes, is there a way to find whether that change is statistically significant with a p-value with a hypothesis of predicting a decrease in the total volume for each month? Or, would I need to use more advanced econometric techniques to identify if there is a significant difference in volumes for an item between months.

    Thank you in advance.

  • #2
    You can do it through regression. You define dummies for the periods that you want to study, and then you regress your volume on those dummies.

    Comment


    • #3
      If I have a pre-period made up of 24 months (2018 and 2019) for each month and then I want to compare changes for each month after COVID-19 for example. Would I just then also include one dummy for each month following that period.

      Code:
      reg volume i.month_num
      Wouldn't degrees of freedom be an issue if I have one observation per month? I'm just trying to determine how best to structure the data to fit a model.

      Comment


      • #4
        Rather, if I want to compare the pre-period (2018 and 2019) in monthly data to individual months in the post-period, how would I approach this? Do I need to create a dummy variable for the entire pre-period then regress each month from the post-period. For example, I want to compare the avg. volume in the pre-period to individual months in the post-period.

        Comment

        Working...
        X