# Thread: 50\$\$\$ to the person who can help me answer 3 questions.

1. ## 50\$\$\$ to the person who can help me answer 3 questions.

willing to paypal 50\$\$\$ to the person who can help me answer 3 questions.

1, The mean weight of 9 randomly selected 5 years old boys babies at a local area was found to be 40 lbs and the standard deviation was found to be 5 lbs. If the weights of newborn babies is approximately normally distributed

A, Find a 95% confident interval for the average weight of all newborn babies at this hospital

B, if you increase sample size, will the confident interval estimate be wider or narrower? Why?

2,The starting salaries of college instructors have a standard deviation of \$2000. How large sample is needed if we wish to be 95% confident that our sample mean will be within \$500 of the true average salary of al college instructors?

5, An editor wants to estimate average the number of pages in bestselling novels, so that his estimate falls within 20 pages of the true average. Assuming that the standard deviation is 63 pages, how large a sample of bestselling novels is needed to achieve
1) 90% confident 2) 95% confident

2. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

Do the questions have to be answered correctly to receive the 50 bucks?

3. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

Originally Posted by trinker
Do the questions have to be answered correctly to receive the 50 bucks?
problem is... how would (s)he even know it?

4. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

It doesn't say. But I oppose using paypal! Those guys are dicks: http://www.regretsy.com/2011/12/05/cats-1-kids-0/

5. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

well... i guess they're so big they can get away with anything now...

if not paypal, then what can people use? i always use paypal...

6. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

** clears voice **

MAH FELLOW TSers:

I CLAIM THIS THREAD IN THE NAME OF THE LULZ NOW.... WELL, BOTH THE LULZ AND THE QUEEN (just because i like her hats & i'm canadian). SO IT'S NOW OPEN TO RANDOMNESS, CRAZINESS AND ANYTHING PEOPLE WANT...

7. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

so this is code for a modified bootstrapped likelihood ratio test in case anyone out there ever needs to test the significance of a variance component using the lme4 package in R for multilevel/HLM models. i did this for a friend of mine who was having trouble in her class so it's addressed to her but if you change around a few things, anyone can use it. just follow along what i told her:

Code:
``````library(lme4)

### IMPORTS THE DATASET. I GUESS YOU'LL CHANGE THIS
### PATH TO MAKE IT WORK

### WORKS OUT THE DESIGN SETTING, DECLARING AS FACTORS WHAT
### SHOULD BE FACTORS, VARIABLES WHAT SHOULD BE VARIABLES AND
### WRAPS UP EVERYTHING IN A NICE DATAFRAME

Rating <- datum\$Rating
TReport <- datum\$TReport
Target <- as.factor(datum\$Target)
myData <- data.frame(Rating,TReport,Target)

### FOLLOWING YOUR CLASSè EXAMPLE, I THINK THIS IS WHAT'S EXPECTED
### FROM YOU TO DO

nullmod  <- lmer(Rating ~ 1 + TReport + (1 | Target)                         , data=myData)
altmod   <- lmer(Rating ~ 1 + TReport + (1 | Target) + (0 + TReport | Target), data=myData)

anova(nullmod, altmod)

############# BOOTSTRAPPED LIKELIHOOD RATIO TEST ##############

### STEP (1) HERE IS THE DIFFERENCE IN LOG-LIKELIHOODS WE ARE GOING
###          TO BUILD OUR BOOTSTRAPPED ESTIMATE AROUND

diffLogLik <- as.numeric(2*(logLik(altmod)-logLik(nullmod)))

### STEP (2) LET'S SET UP THE DESIGN OF THE BOOTSTRAP. reps IS HOW MANY TIMES
TO RUN IT. lrstat IS WHERE I WILL STORE MY RESULTS. y IS A VECTOR
OF RESPONSES SIMULATED FROM THE NULL DISTRIBUTION

reps <- 100
lrstat <- numeric(reps)
y <- simulate(nullmod)

### STEP (3) HERE'S THE SETTING OF THE BOOTSTRAP

for (i in 1:reps) {
y <- unlist(simulate(nullmod))
null <- lmer(y ~ 1 + TReport + (1 | Target), data=myData)
alt  <- lmer(y ~ 1 + TReport + (1 | Target) + (0 + TReport | Target), data=myData)
lrstat[i] <- as.numeric(2*(logLik(alt)-logLik(null)))
}

### STEP (4) HERE IS WHERE MY RESULTS GET REPORTED. I SHOULD BE LOOKING FOR A
### pVALUE LESS THAN 0.05 AND WITH A SMALL STANDARD ERROR TO BE EXTRA-SURE
### IT DOESNT CROSS OVER THE 0.05 DEATHZONE!!!

pValue <- mean(lrstat > diffLogLik)
pValueSTDError <- (pValue*(1-pValue))/reps

pValue
pValueSTDError``````

9. ## The Following 2 Users Say Thank You to TheEcologist For This Useful Post:

spunky (12-06-2011), trinker (10-02-2012)

10. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

Way to go guys. Once again you all ruined my capitalistic scheme and no early Christmas money. Though we have no rescind of offer this time...

I swear some contributor is making fake member names and getting on here and generating fun little spinoff threads for the members.

*Wait*

....

Dason?????????????

11. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

Originally Posted by karkid
1, The mean weight of 9 randomly selected 5 years old boys babies at a local area was found to be 40 lbs and the standard deviation was found to be 5 lbs. If the weights of newborn babies is approximately normally distributed
Come on, 5 year old babies...
That's a picture of the Italian Stallion when he was an 8 year old baby

12. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

Originally Posted by trinker
Way to go guys. Once again you all ruined my capitalistic scheme and no early Christmas money. Though we have no rescind of offer this time...
OP was only going to give you 50 \$ signs. Already got 3 of them in the thread title. \$50 on the other hand...

13. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

Originally Posted by Mean Joe
OP was only going to give you 50 \$ signs. Already got 3 of them in the thread title. \$50 on the other hand...
Very perceptive that karkid or should I say Dason, is a crafty fellow.

14. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

Originally Posted by trinker
I swear some contributor is making fake member names and getting on here and generating fun little spinoff threads for the members.

*Wait*

....

Dason?????????????
Hey now... I already used my extra email account to make spunky think twice about spouting out calls for certain authors deaths. And I'm way too lazy to go through the process of getting another email account and creating yet another sockpuppet just for some laughs. OR AM I?!?

Also - I adamantly deny any accusations that I have created any sockpuppet accounts in the past. Ignore the implications in the first paragraph that I may have done exactly this. The person that wrote that first paragraph can not be trusted.

15. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

Originally Posted by Dason
to make spunky think twice about spouting out calls for certain authors deaths.
it was for the book, not the authors!!! the booooooook!!!! *slow-mo mode* theeeeee boooooOOOOOooooOOOOoooKKKK...

ok, so lemme rephrase it then:

DEATH TO Using Multivariate Statistics (Hardcover) BY Tabachnick & Fidell!!!!

16. ## Re: 50\$\$\$ to the person who can help me answer 3 questions.

****... that might be too long for a user name...