Thoughts While Checking My Language Skills

I had a post idea all lined up and then something new landed in my lap. Well, more accurately, it landed in my inbox. I have been tasked with providing some aggregates and means for certain service calls. When I’m looking through some collected data to give me the baseline, I realized that I have four separate columns in my dataset that could be used as the starting timestamp. Many of the rows in the dataset have identical values in all 4 columns. For the rows that don’t, there can be differences between columns of up to 45 seconds. When I asked the requester if there was a preference for which column to use and I received explanations about how further research should be left for operations. I’m ok with that. Ultimately, I was able to learn that there was no preference and I am free to select the starting point I wish.

However, it did get me thinking. I want to know if there is a specific timestamp that should be used as the starting point? Can that be determined statistically? I know that I can create SQL to isolate the earliest time point when there is a discrepancy. However, this could be an interesting project. So, I think I’m going to design it and work on it a bit.