Dekoyl's question thread

[dekoyl]

Ah thanks shinjitsuzx. I did everything you did except I worked the surd bit wrong. Stupid mistake.
-------------------------------------------------------------------------------------------------------------------
Anyone know the remaining question? I just ripped the paper with my rubber after erasing it 10 times.

2. if
The above is a probability density function. What is the variance of X?
I'm sure I'm doing everything right but I keep getting .

[ReVeL]

I just got using the calculater. :/

[Glockmeister]

...2 unknowns and 1 piece of info? How's this work. Are you sure that p isn't a typo <_<

I was doing the question over breakfast... I was wondering what the P was.

[dekoyl]

...2 unknowns and 1 piece of info? How's this work. Are you sure that p isn't a typo <_<

I was doing the question over breakfast... I was wondering what the P was.

Yeah sorry about that It was indeed a typo. The 'p' was meant to be a '9'. I was reading over the previous question about the thing and I guess the 'p' just got stuck in my head.

[dekoyl]

Another one..




Find the value of to 3 decimal places such that the graphs and touch each other once only.

Thanks.

[/0]

Another one..




Find the value of to 3 decimal places such that the graphs and touch each other once only.

Thanks.

I'm assuming 'touch' means a tangency at a point. (Which I guess would be necessary for one intersection only)





At the point where and   touch, they must have the same gradient i.e.



Also, they must have the same y-value

(as the solution to must lie on the line )

So







If ,

Solve



If

Solve



Hmmmm, something interesting to note, gives a 'touch' where the gradient and the value are the same, but the two graphs still cross!

[dekoyl]

Thanks heaps /0.
That was great - just a side note the answer only wanted a = 2.156.

[dekoyl]

What wording indicates that you round up/down with answers to probability questions?

It's kind of hard to explain what I'm asking but I'll try.
For example, if you end up with an answer of 17.68 apples, the probability of 18 isn't as "sure" as getting 17.
If it said to the nearest whole number would it be 18? And if it said to the nearest apple would it be 17?

Thanks.

[bec]

If you're talking about things like PEOPLE and that you can't have half of, you always need to round up. For example, if you do all the calculations and find that "for a 20% chance that someone has brown eyes in the class, there must be 14.2 girls in the class", then you need to say 15. In this case, 14 girls wouldn't be enough, you can't have 14.2 girls, and 15 is just right...this doesn't just happen in probability either, but also in logs and exponentials (no examples are coming to mind, but half life questions usually seem to have this 'dilemma')

I'm not sure - someone back me up on this - but I think that if they say to round to the "nearest whole number", you have to do just that. So for the 14.2 girls example, you'd say 14. I think most questions we get will be pretty clear though.

Hope that helps

[Glockmeister]

My opinion is that you always round down when it refers to people so that you don't arbitary make .5 people.

If it says "nearest whole number" you follow the standard rounding principles; that is.

for numbers x.0 to x.4 where x is an integer, you round down.
for numbers x.5 to x.9 where x is an integer, you round up.

[Mao]

I think for probability [specifically for means and other measurements of central tendency], the rule of "NO HALF OF HUMANS" doesn't really apply.

e.g. if there was a question on workers drinking coffee/tea in an office, and you get an average of 5.3 people drinking coffee, there is no need to panic ["OH NO THE 0.3 PERSON IS STEALING MAH COFFEE"], the 5.3 is left as is.

[or so I've been told]

[dekoyl]

Thanks for the various suggestions. I'll further consolidate with my teacher.

And also...
How does one sketch the integral or derivative of a function on a Ti-84 calculator?
I seldom use the calculator for these and I've forgotten (not sure if it's even possible.)

Thanks.

[Glockmeister]

Dervive the function by hand and then put it in the calculator...

[shinny]

Don't think it's possible directly since it can't do algebra =\

As for the rounding, from what I see, it depends on what type of question its asking. If its asking for a mean/expected value, don't round as it's asking for an actual STATISTICAL FIGURE (i.e. don't round on these). However, if they ask how many trials would you need to guarantee at least a .6 probability of success, then you have to round up even if its 14.000001 because at 14, you still wouldn't have hit .6, and you can't have a fraction of a trial.

[Mao]

ummm

it's possible.

for TI-84 series and TI-83+ (dunno about the other ones)
the following functions are probably found in MATH

integral of a function [graph starting at (0,0)]

Code: [Select]

y1=fnInt(...function here...,X,0,X)

and

derivative of a function

Code: [Select]

y1=nDeriv(...function here...,X)