eq: 2.22 subscript missing for y_i (digital edition, cannot confirm for printed)

[keblu]

Small subscript missing in equation 2.22

In the digital edition (maybe printed also) we have:

lhd = Πⁿ_i=1 p(y_i | x_i) =Πⁿ_i=1 p_i^yi(1-p_i)^1-y_i

should be:

lhd = Πⁿ_i=1 p(y_i | x_i) =Πⁿ_i=1 p_i^y_i(1-p_i)^1-y_i

Note the added subscript at p_i^y_i instead of p_i^yi

[keblu]

Realized that there is a couple of those in the kindle edition, I guess that this won't be the case in the printed one.

[mataddy]

Thanks keblu! Where else did you catch these?

[keblu]

1. paragraph right after eq. 2.30, for var(a'ε) = a[X] var(ε) a, I guess [X] is for '.

2. paragraph right after that you have diag([σ²₁,σ²₂,...,σ²n]) instead of diag([σ²₁,σ²₂,...,σ²_n])

3. paragraph after eq. 2.31 you have σ²i instead of σ²_i

4. Seems like an error here in Section Space and time at the beginning you have:

log(y_t) = α + βt_t + βm_t + εt

in that case there is two time β without subscript (indicating that this is the same coefficient), βt_t is a bit strange since obviously t at index t will be t, it should probably be simply βt, and εt should be ε_t.

5. Paragraph where you have: A more subtle issue is that of autocorrelation .

A bit after that you have:

The errors appear to be correlated over time, such that εt is related to εt_-1. That relationship violates our basic independence assumption, which holds that εi ⟂⟂ ε_j for all i ≠ j.

should probably be:

The errors appear to be correlated over time, such that ε_t is related to ε_t-1. That relationship violates our basic independence assumption, which holds that ε_i ⟂ ε_j for all i ≠ j.

6. Section where it is written: It is also possible to extend the idea of the AR to higher lags

AR(p): Y_t= β₀+ β₁Y_t-1 + ... + + β_pY_t-p + ε.

there is a missing subscript on ε, should be ε_t

7. Just before eq. 3.1 :

suppose you have data [x₁,y₁]...[xn,y_n]

should be:

suppose you have data [x₁,y₁]...[x_n,y_n]

8. before eq. 3.4

λ₁> λ₂... > λT should probably be: λ₁> λ₂>... > λ_T

For now that is about it. Again, I don't have the printed edition so I am not sure if some of those are because of the Kindle app.

[mataddy]

Thanks! Massively helpful. If you find more please put them here. Subscript errors were a nightmare in production and we obviously didn’t catch them all. Appreciate the edits. Hope you like the book!