Would love any direction on how to get to the finish line. I'd love to be able to decipher what tusharm is describing in this post to the point where I could replicate this myself. I've tried exponential smoothing, moving average, and even the forecast function, but all of them don't seem to produce the result that seems to make as much sense as the power curve. Example 1 : Lily is getting trained for a 10K race. I need to forecast the rest of the values through 48 months. We can use two points on a trend line to write an equation in slope-intercept form for the trend line.
I have the one column representing decreasing transactions each month that look like this: The power curve trendline seems to have the closest R-square value (again still trying to learn what that is) to 1 and well, it just looks closer than any of the other trendlines. Anyway, I am doing a lifetime value analysis and my data set shows a series of transactions starting off strong and steadily declining over time. I am cursed with desperately trying to catch up from not taking math or statistics in college. OK, I am trying to solve the exact same problem that chastst is, except I am dumber, well, a lot less smart.
0 kommentar(er)
