My original process had two scalars because the estimation file consisted of only two points in time (2019 and 2020). The new estimation file has 16 points in time, one for each month. I think the best strategy is to have 16 scalars then, because the QCEW data is available monthly. Even the months in 2018 where LEHD is available need the data to be scaled to the specific month it represents.
My issue is that it is not clear to me what the base of the scalar should be, or what my starting point should be. I will be using the 2018 LEHD data because it is the most up-to-date data that is available. My first instinct is to average the QCEW data across all months in 2018. Does that sound right to you?
My original process had two scalars because the estimation file consisted of only two points in time (2019 and 2020). The new estimation file has 16 points in time, one for each month. I think the best strategy is to have 16 scalars then, because the QCEW data is available monthly. Even the months in 2018 where LEHD is available need the data to be scaled to the specific month it represents.
My issue is that it is not clear to me what the base of the scalar should be, or what my starting point should be. I will be using the 2018 LEHD data because it is the most up-to-date data that is available. My first instinct is to average the QCEW data across all months in 2018. Does that sound right to you?