database feature request.
#2
Tom,
It's been requested before and once I get my major features out of the way, I'll stick it on my Todo list. You'll see in the coming days, however, that I have been working hard on the latest and greatest of GreenHybrid.
It's been requested before and once I get my major features out of the way, I'll stick it on my Todo list. You'll see in the coming days, however, that I have been working hard on the latest and greatest of GreenHybrid.
#5
Re: database feature request.
Hey Jason, I'm back, looking for more ways to chop up the wonderful mileage DB for our edification :-)
Could you normalize a model's average mpg to miles driven ?
E.g., if one Prius had been driven 100 miles at 50 mpg, and the second Prius 200 miles at 60 mpg, the normalized mean for the two car group would be 170/3 mpg. Doing so would give more weight to cars that had been on the road longer, and so emphasize driver experience and seasonal averaging.
Could you normalize a model's average mpg to miles driven ?
E.g., if one Prius had been driven 100 miles at 50 mpg, and the second Prius 200 miles at 60 mpg, the normalized mean for the two car group would be 170/3 mpg. Doing so would give more weight to cars that had been on the road longer, and so emphasize driver experience and seasonal averaging.
#6
Re: database feature request.
Eric, I thought about this quite some time ago and concluded that it was an inaccurate and misleading way to report data. While this would be ideal in controlled circumstances, the fact that our sampling is NOT controlled means that significance of a car's data depends solely on distance traveled. This doesn't take into account the variety of variables (temp, trip time, speed, etc.) that should also be taken into consideration. As a result, we'd end up with a figure, but it won't be usable.
The database currently reports the median reported lifetime mileage from a voluntary sample of hybrid owners. Say some crazy driver drove 1,000,000 miles in -20 degree weather for 3 minutes at a time. No big deal.
What result would normalizing bring? Say that same crazy driver was normalized. He'd skew the average by gigantic proportions, lowering the average mileage and influencing the overall figure when he really ought not to. His temperature and trip time are NOT related to what REAL hybrid owners drive.
The database currently reports the median reported lifetime mileage from a voluntary sample of hybrid owners. Say some crazy driver drove 1,000,000 miles in -20 degree weather for 3 minutes at a time. No big deal.
What result would normalizing bring? Say that same crazy driver was normalized. He'd skew the average by gigantic proportions, lowering the average mileage and influencing the overall figure when he really ought not to. His temperature and trip time are NOT related to what REAL hybrid owners drive.
#7
Re: database feature request.
Every data analysis done in good faith has strengths and weaknesses. As you point out, a mileage normalized result would appear skewed to an average driver if a high mileage extreme driver was in the mix. Since I think that scenario less likely than a result skewed by a bunch of drivers who record data for a couple of tanks in one season and then stop, the proposed analysis has merit.
And in any case, I am not suggesting one or the other, but multiple views that help point out mileage performance obscured by our as yet relatively immature data. Incidentally, since I think the data will turn out to follow a normal (gaussian) distribution, over time the mode and median will converge, as will a result that has been calculated by mileage normalization.
Your comments regarding trip time and temperature are spot on. However, if you have 5 variables, taking 2 into account is better than 1 .. 3 is better than 2 ... etc, assuming equal weights.
And in any case, I am not suggesting one or the other, but multiple views that help point out mileage performance obscured by our as yet relatively immature data. Incidentally, since I think the data will turn out to follow a normal (gaussian) distribution, over time the mode and median will converge, as will a result that has been calculated by mileage normalization.
Your comments regarding trip time and temperature are spot on. However, if you have 5 variables, taking 2 into account is better than 1 .. 3 is better than 2 ... etc, assuming equal weights.
#8
Re: database feature request.
Noted. I'll see about what I can do in the future about this. There are obviously TONS of things that can be done. There are just a rediculous amount of opportunities. Perhaps this is one best saved for *gasp* RHMDB 2.0
Thread
Topic Starter
Forum
Replies
Last Post