Dear All,
I am doing an analysis and I use Excel for drawing charts, I want to compare trendlines of two different charts. Lets say we have two charts: 1) cost vs. drilling time and 2) cost vs. water depth. I put all the data and drew 2 charts (scattered) and drew trendlines (linear) for them also. Now I want to see which one of these parameters (drilling time or water depth) can affect the cost more. First I said, I get slope of the trendlines and the steeper is the most effective one. But I came to this point that data range for horizontal axes is not the same, in 1st case the raneg is (0-5000) and in the 2nd case the range is (0-400). Obviously, the slope for the 1st one is smaller but it does not say that the 2nd parameter is more effective than the 1st one. So, I want to know that is there any other way that I can compare these two trendlines and say which one is steeper than the other (more effective)? I need to a kind of normalize these trendlines in order to compare them with each other. I would appreciate your comments and helps.
Hooman
Welcome to the forum.
First, in both cases the independent variable is a very weak predictor of cost (R2 = 7% and 2% respectively).
Second, if you're going to use multi-variate regression, I think you need one dataset with depth, drilling time, and cost.
Microsoft MVP - Excel
Entia non sunt multiplicanda sine necessitate
Thanks to shg, I see your point but I think still the trendlines are showing something and a relation between parameters (drilling time, water depth) with (cost) although it is a poor relation coz you can see we have some data which are really far from other values. Now I made a new parameters which make a kind of normalization in our parameters (drilling time, water depth), I devided these parameters by 5000 and 400 respectively to convert them into one range (0-1) which I think we call it normalization, and if now you check the edited attached file, you can see that both graphs have the same horizontal and vertical axes range which makes new comparison of the trendlines sensible and you can say that slope of water depth curve is bigger than the slope of drilling time and therefore water depth influence the cost more than drilling time. What do you think? any comments? Thanks for your time.
My sense is that you cannot combine two datasets derived from different sets of observations, combine them, and draw meaningful conclusions. The wells drilled in the two sets of data may have had very different characteristics, such that combining them makes no sense.
I guess it depends on what you actually know about the data.
Microsoft MVP - Excel
Entia non sunt multiplicanda sine necessitate
There are currently 1 users browsing this thread. (0 members and 1 guests)
Bookmarks