I got .083 also…
cov is /n-1. the n-1 term cancels out and corr=cov(x,y)/stdev(x)*stdev(y)
that one was for sure .339
isnt the covariance formula Cov= Sum of N( x-meanX)(y-Ymean) / N-1…I see on the net all forms for N as in N, N-1 and N-2…
it is. and the formula for stdev also has n-1 in the denominator. so they cancel out
anyone remember the sum(x2), sum(y2) and sum(x-x0)(y-y0)? cov cancelled out (n-1) or (n-1)square?
I thought Cov was given :S. I ended up with 0.690 but you’re probably right. Can’t believe I f*cked that up.
COncerning the hypothesis question: They asked what the right Ho was. I choose Bo = 0, because he needs to reject the Bo. who agrees. If I am correct I have a 6/6 for quant…
I got .083 for correlation as well.
mahsun23 Wrote: ------------------------------------------------------- > COncerning the hypothesis question: > > They asked what the right Ho was. I choose Bo = 0, > because he needs to reject the Bo. > > who agrees. If I am correct I have a 6/6 for > quant… >1 IS CORRECT
REgarding the n-1s in denominator…they don’t cancel out, do they? you divide the numerator by n-1 which is 34 (?) and then EACH denominator by 34 and multiply them. Am i dead wrong (i haven’t seen anyone else who did it this way)? i took the number that looked like the cov, that is sum(x-x)(y-y) and divided by 34…then divided that by the product of the square root of the two givens dived by 34 each…so, i can’t remember my exact answer but the math is: (sum(x-x)(y-y)/34) / ((sqrt((x-x)^2)/34)(sqrt(y-y)^2)/34))…anyone have anythoughts on this? Basically, i couldn’t remember which one of those numbers was SSE or RSS or what have you, so i tried to replicate the corr= cov/ sigma*sigma…and i felt like the numbers that were given needed to be converted to cov by dividing by n-1, and to sigma by dividing by n-1 and then taking square root of each before multiplying.
only way to be sure if we can remember the sum of square given in the question.
It was something like 70