View Full Version : Statistics Question for any Math Whizzes
RandallMarshall
01-21-2011, 02:59 AM
I'm pretty sure I understand "Mean Absolute Deviation", but this problem won't work for me. I have to use the 2-day moving average which is the last two numbers... 88 and 90. Even then, I still can't get the right number. I've tried everything you can imagine. I'm 100% certain the answer is 3.20, but I can't figure out how to get it. If anyone could show me a step by step walk through I will rep you beyond your wildest dreams. :-P
Daily high temperatures in St. Louis for the last week were as follows: 92, 91, 94, 95, 96, 88,90.
The mean absolute deviation based on a 2-day moving average = 3.20
Count_Chocula
01-21-2011, 04:55 AM
pie are round, cornbread are squared
Count_Chocula
01-21-2011, 05:00 AM
go ask goodwill hunting
goodwill ain't open this early, try the homeless shelters
Count_Chocula
01-21-2011, 05:03 AM
or op could do what the nazis do and just burn the math book
those were books on projective geometry, not statistics
emelvee
01-21-2011, 05:36 AM
:look:
veritableqndry
01-21-2011, 06:06 AM
:look:
You're an absolute mean deviation, do you know that? :annoyed:
iconian
01-21-2011, 06:08 AM
I'm pretty sure I understand "Mean Absolute Deviation", but this problem won't work for me. I have to use the 2-day moving average which is the last two numbers... 88 and 90. Even then, I still can't get the right number. I've tried everything you can imagine. I'm 100% certain the answer is 3.20, but I can't figure out how to get it. If anyone could show me a step by step walk through I will rep you beyond your wildest dreams. :-P
Daily high temperatures in St. Louis for the last week were as follows: 92, 91, 94, 95, 96, 88,90.
The mean absolute deviation based on a 2-day moving average = 3.20
is that based on the bell curve? if so, you need to calculate it...you can't just guess...
I suspect absolutely mean deviation maybe at 68%? then you have to do it. I think there is a way to doit via t-83 calculator and stuff... But I hate stats
Putz1103
01-21-2011, 07:49 AM
I got 3.57. That's the best I could do.
Take the rolling average, in this case the average of 90 and 88: 89. Then find the deviation of every day from that average. Absolute value the deviations. Then average the deviations. That got me to 3.57. I'm not sure how you are 100% sure it's 3.2 if you don't know how to get to that value, but good luck anyway.
ikrit
01-21-2011, 08:06 AM
I got 3.57. That's the best I could do.
Take the rolling average, in this case the average of 90 and 88: 89. Then find the deviation of every day from that average. Absolute value the deviations. Then average the deviations. That got me to 3.57. I'm not sure how you are 100% sure it's 3.2 if you don't know how to get to that value, but good luck anyway.
got the same answer.
sometimes the answer in the back of the book is wrong :P
RandallMarshall
01-21-2011, 12:04 PM
is that based on the bell curve? if so, you need to calculate it...you can't just guess...
I suspect absolutely mean deviation maybe at 68%? then you have to do it. I think there is a way to doit via t-83 calculator and stuff... But I hate stats
There's no need to base this off a curve. I've actually tried a few tutorials for my TI-84.
I got 3.57. That's the best I could do.
Take the rolling average, in this case the average of 90 and 88: 89. Then find the deviation of every day from that average. Absolute value the deviations. Then average the deviations. That got me to 3.57. I'm not sure how you are 100% sure it's 3.2 if you don't know how to get to that value, but good luck anyway.
got the same answer.
sometimes the answer in the back of the book is wrong :P
Well the thing is I can get free examples that give me the answers, but they won't tell me the solution. Putz, that would be the 3-day moving average. I swear I know how to do this, but I can't get the right answer. I've attached a screen shot that includes some of the other questions. You can see it includes the answers.
RandallMarshall
01-21-2011, 12:58 PM
I finally got it thanks to someone helping me!
Numbers - 92, 91, 94, 95, 96, 88,90
2-Day Averages - 91.5, 92.5, 94.5, 95.5, 92
Absolute Error - 2.5, 2.5, 1.5, 7.5, 2
MAD = 3.2
Putz1103
01-21-2011, 02:38 PM
I finally got it thanks to someone helping me!
Numbers - 92, 91, 94, 95, 96, 88,90
2-Day Averages - 91.5, 92.5, 94.5, 95.5, 92
Absolute Error - 2.5, 2.5, 1.5, 7.5, 2
MAD = 3.2
Oh that's garbage. You take the deviation of the third day based on the average of the previos two? That could be done so many different ways, how were you to know that was the method they wanted? Boo to statistics...
cgrady
01-21-2011, 03:23 PM
But only one of those ways is the right way :)
Putz1103
01-21-2011, 03:40 PM
But only one of those ways is the right way :)
Bull! There is no spoon!
VorlonFrog
01-21-2011, 04:29 PM
Are you using "n" or "n-1" for your standard deviation derivation?
RandallMarshall
01-21-2011, 04:48 PM
Are you using "n" or "n-1" for your standard deviation derivation?
It's not standard deviation so you only use "n". It's similar.
emelvee
01-21-2011, 06:11 PM
You're an absolute mean deviation, do you know that? :annoyed:
I do! :woot: