Errors Found in the Class Notes

p. 2-6, 16th line

in the 2nd integral, the square brackets of the integrand should be raised to the k-1 power (as is the case for the integrand of the first integral on that line)

p. 2-6, 22nd line

in the joint pdf, the last set of square brackets should be raised to the n-s power (as is the case in expression (2.6.8) at the bottom of p. 45 of the text)

p. 2-31, 9 lines from bottom

the z in front of the 1.9600 should have a subscript of 0.025 (instead of 0.05)

p. 3-12, 9 lines from top

after "exceeds 0.44 for n >= 10" but before the period at the end of the sentence insert the following: and approaches, but never exceeds, 0.5 as n tends to infinity

p. 3-14, 4 lines from top

after "of the longest run," the next word should be one instead of or

p. 3-19, 14 lines from top

the 5th to the last word on the line is "broadness" (my handwriting isn't too good there)

p. 3-19, 15 lines from top

right before the word "alternative" insert the word the

p. 3-27, 4 lines from top

instead of "greater for" it should be greater tendency for

p. 4-5, bottom line

N_i should be inserted right after the second summation

p. 4-6, 2 lines from bottom

the expectation in the denominator of the last expression for Q should have a subscript of 0 (like the expectation in the numerator does)

p. 4-22, 7 lines from bottom

X₍₂₎ should be X₍₃₎

p. 4-28, 17 lines from bottom

p. 576 should be p. 565

p. 4-33, 7 lines from top

"contained" should be obtained

p. 5-10, 1st line below the table

change "for the sign test" to for the power of the sign test

p. 5-16, 3rd line

there should be a } after n(n+1)/2 to close off the set of integers

p. 5-27, 7 lines from bottom

after "are the signed random scores" and before the period, insert

, if we're testing for a treatment effect using matched-pairs differences, or doing a test about a median with η₀ = 0

p. 6-5

at the top of the page, above the figure, I meant to put θ > 0 to indicate which case was sketched

p. 6-15, 3rd line of 1st paragraph

"rejects" should be reject

p. 6-15, 2nd sentence of last paragraph

delete everything before "then the prob. that V" and change "then" to Then

p. 7-2, last line

there are only three = signs on the last line ... some stray marks on the page make two minus signs look a bit like = signs

p. 7-5, 7th line

for the 1st of the 3 sums on the line, the sum is for i = 1 to N, for the 2nd of the 3 sums on the line, the sum is for i = 1 to N - 1, and for the 3rd of the 3 sums on the line, the sum is for j = i + 1 to N

p. 7-8, 9th & 10th lines

an = sign is missing about 60% of the way across the double line display

p. 7-9, 12 lines from bottom

the two terms being added look the same, but the 1st z vector should not have an accent

p. 7-9, 5 lines from bottom

the vector of 0s and 1s should be the reverse order of the one above it (7 lines from bottom) ... so 5 lines from the bottom it should be 0, 0, 1, 1, 1, 0, 1, 0, 0, 1

p. 7-12, 15th line

"less general to" should be less general than

p. 7-13, 5th line from bottom

"an" should be and

p. 8-1, 3rd line from bottom

"you can get" should be you can with high probability get

p. 8-10, 5th line

the term being summed should be multiplied by Z_i

p. 8-11, 3rd and 4th lines from bottom

scratch through my parenthetical Note, since I was wrong about the book being wrong ... in addition to what I have as being equivalent to the W-M-W test for N odd, if N is even then the book's prescribed scores of -(N-1)/2, -(N-3)/2, ..., -3/2, -1/2, 1/2, 3/2, ..., (N-3)/2, (N-1)/2 also create a test equivalent to the W-M-W test

p. 9-1, line right under specification of H₁

"two-sided" should be one-sided

p. 9-6, 10th line from bottom

replace M_N by

the Freund-Ansari-Bradley test statistic (which is denoted by F_N in the text)

p. 10-2

in two places where I give the numerator of the F statistic, I left out the factor n_i

p. 10-6, 4 lines above the sketches at the bottom of the page

there should be a ) after the word equal (to close off the open parenthesis before ending the sentence)

p. 10-19, 5th line from bottom

there should be a + before the first term on the line (since the sum continues from above)

p. 10-20, 8th line from top

in the formula for the variance, the n_i shouldn't be squared

p. 10-24, 4th line from top

the word "test" at the end of the line can be deleted since it's also the first word on the next line (and it shouldn't be there twice)

p. 10-27, 5th and 6th lines from top

instead of writing the alternative hypothesis as I did, it'd be better to just put

against the general alternative that at least one of the treatment distributions is stochastically larger than the control distribution

p. 10-27, 13th and 14th lines from top

instead of writing the alternative hypothesis as I did, it'd be better to just put

against the general alternative that at least one of the treatment distributions is stochastically smallerer than the control distribution

p. 10-27, last line

the 2nd word on the line should be because instead of "be cause" and the 2nd to the last word should be But instead of "Bul"

p. 10-28, 7th line

the last thing on the line should be a binomial coefficient ... it should not be a fraction in parentheses

p. 11-3, 3rd line

the first event in the last intersection on the line should be { X_i > X_j } instead of { X_i > Y_i }

p. 11-12, last line

there should be a ) at the end of the line (to close off the parentheses opened on the 4th line from the bottom)

p. 12-15, 2nd line from bottom

add the word null to the end of the line (since it's the mean and variance of W under the null hypothesis that is being referred to)