Total Posts:14|Showing Posts:1-14
Jump to topic:

My thoughts on number rounding

dylancatlow
Posts: 12,245
Add as Friend
Challenge to a Debate
Send a Message
4/20/2013 11:12:47 AM
Posted: 3 years ago
Whenever an accurate figure is desired, and rounding is necessary, wouldn't rounding a non-siginifcant figure to five rather than zero better reflect the true value? For instance, say I have a scale and I want to weigh myself. The scale is only accurate to the nearest 10 pounds. I weight myself and it spits out 170 pounds. Because it's accurate only to the nearest ten pounds, I could also weigh 171, 172, 173, 174, 175, 176, 177, 178 or 179 pounds. Because each of these is equally likely, and a single number needs to represent these possibilities, wouldn't it make sense to take the average of them, and choose the statistically 'least wrong' number, e.g. 175?
Subutai
Posts: 3,197
Add as Friend
Challenge to a Debate
Send a Message
4/20/2013 11:43:58 AM
Posted: 3 years ago
That scale would read 170 if you weighed 165-175 pounds, not 170-180.
I'm becoming less defined as days go by, fading away, and well you might say, I'm losing focus, kinda drifting into the abstract in terms of how I see myself.
dylancatlow
Posts: 12,245
Add as Friend
Challenge to a Debate
Send a Message
4/20/2013 11:52:17 AM
Posted: 3 years ago
At 4/20/2013 11:43:58 AM, Subutai wrote:
That scale would read 170 if you weighed 165-175 pounds, not 170-180.

It depends on the kind of scale. If the scale just compared your weight to 10 pound increments, and indicated at which point you weighed the same or more than 170 but not more than 180, then what I'm saying would be relevant.
F-16_Fighting_Falcon
Posts: 18,324
Add as Friend
Challenge to a Debate
Send a Message
4/20/2013 1:22:27 PM
Posted: 3 years ago
At 4/20/2013 11:52:17 AM, dylancatlow wrote:
At 4/20/2013 11:43:58 AM, Subutai wrote:
That scale would read 170 if you weighed 165-175 pounds, not 170-180.

It depends on the kind of scale. If the scale just compared your weight to 10 pound increments, and indicated at which point you weighed the same or more than 170 but not more than 180, then what I'm saying would be relevant.

Then, it would be an unusually sucky scale.
Andromeda_Z
Posts: 4,151
Add as Friend
Challenge to a Debate
Send a Message
4/21/2013 1:19:03 AM
Posted: 3 years ago
No, in that case it would be best to say 170 (" 10) pounds. This shows both the reading and the amount of uncertainty. Well, not really, because the last digit is just so completely uncertain that you should use scientific notation and drop it entirely, but you get the idea. The scale gave you a reading and that reading is useful. But, so is the uncertainty.
Subutai
Posts: 3,197
Add as Friend
Challenge to a Debate
Send a Message
4/21/2013 3:14:03 PM
Posted: 3 years ago
At 4/20/2013 11:52:17 AM, dylancatlow wrote:
At 4/20/2013 11:43:58 AM, Subutai wrote:
That scale would read 170 if you weighed 165-175 pounds, not 170-180.

It depends on the kind of scale. If the scale just compared your weight to 10 pound increments, and indicated at which point you weighed the same or more than 170 but not more than 180, then what I'm saying would be relevant.

Then that scale doesn't really know rounding in the first place. If it goes by ten-pound increments, then a particular scale reading should encompass weights from 5 pounds below the reading to 5 pounds above the reading, not solely 10 pounds above the reading.
I'm becoming less defined as days go by, fading away, and well you might say, I'm losing focus, kinda drifting into the abstract in terms of how I see myself.
R0b1Billion
Posts: 3,732
Add as Friend
Challenge to a Debate
Send a Message
4/21/2013 10:03:51 PM
Posted: 3 years ago
Rounding is important for scientific calculations, where a failure to do so would represent a misleading level of accuracy for your calculations. If you weight 150 pounds, you don't want to insinuate it's necessarily 150.0
Beliefs in a nutshell:
- The Ends never justify the Means.
- Objectivity is secondary to subjectivity.
- The War on Drugs is the worst policy in the U.S.
- Most people worship technology as a religion.
- Computers will never become sentient.
darkkermit
Posts: 11,204
Add as Friend
Challenge to a Debate
Send a Message
4/21/2013 10:07:11 PM
Posted: 3 years ago
Generally if you say that something is off, you use +/-. So it would be 170 +/- 10 pounds.

Also, the degree of accuracy is give in standard deviations. SO one standard deviation can be 5 pounds, so its 170 +/- (two standard deviations) +/-10 pounds.
Open borders debate:
http://www.debate.org...
Andromeda_Z
Posts: 4,151
Add as Friend
Challenge to a Debate
Send a Message
4/21/2013 10:31:28 PM
Posted: 3 years ago
At 4/21/2013 1:19:03 AM, Andromeda_Z wrote:
No, in that case it would be best to say 170 (" 10) pounds. This shows both the reading and the amount of uncertainty. Well, not really, because the last digit is just so completely uncertain that you should use scientific notation and drop it entirely, but you get the idea. The scale gave you a reading and that reading is useful. But, so is the uncertainty.

Ugh 17 +/- 10 pounds. I didn't realize that got messed up until now...
darkkermit
Posts: 11,204
Add as Friend
Challenge to a Debate
Send a Message
4/21/2013 10:36:43 PM
Posted: 3 years ago
At 4/21/2013 10:31:28 PM, Andromeda_Z wrote:
At 4/21/2013 1:19:03 AM, Andromeda_Z wrote:
No, in that case it would be best to say 170 (" 10) pounds. This shows both the reading and the amount of uncertainty. Well, not really, because the last digit is just so completely uncertain that you should use scientific notation and drop it entirely, but you get the idea. The scale gave you a reading and that reading is useful. But, so is the uncertainty.

Ugh 170 +/- 10 pounds. I didn't realize that got messed up until now...

fixed. You still messed it up :p
Open borders debate:
http://www.debate.org...
Andromeda_Z
Posts: 4,151
Add as Friend
Challenge to a Debate
Send a Message
4/21/2013 11:52:45 PM
Posted: 3 years ago
At 4/21/2013 10:36:43 PM, darkkermit wrote:
At 4/21/2013 10:31:28 PM, Andromeda_Z wrote:
At 4/21/2013 1:19:03 AM, Andromeda_Z wrote:
No, in that case it would be best to say 170 (" 10) pounds. This shows both the reading and the amount of uncertainty. Well, not really, because the last digit is just so completely uncertain that you should use scientific notation and drop it entirely, but you get the idea. The scale gave you a reading and that reading is useful. But, so is the uncertainty.

Ugh 170 +/- 10 pounds. I didn't realize that got messed up until now...

fixed. You still messed it up :p

Well, this is awkward...