---

Page 2 Table 1.1 has labels out of alignment.

Page 38 Apparently the Nadaraya-Watson kernel estimate is now also known as a general regression neural network (GRNN).

Page 52 In Table 1.9 the x₅ splits with j = 4, 10 should not be considered since j = 5, 11 have duplicate predictor variables.

Page 53 In Table 1.10 the x₅ split with j = 10 should not be considered since j = 11 has a duplicate predictor variable.

Page 135 Just before the display AFTER (4.2.2) there should be a minus sign in the formula for the derivative of V inverse.

The subsequent display is correct because the missing minus sign would appear on both sides of the equation.

Page 144 In the display AFTER (4.5.1) there should not be a transpose (prime) on the extreme right.

Page 162 In the third line of Example 5.1.1, \gamma is a column vector not a row vector i.e.,

put a prime on one side of the \gamma equality.

Page 167 "Murry Clayton" should be "Murray Clayton."

Page 174 Fourth line, (I-M_Z)Y - should be (I-M_Z)Y +

Page 175 Ninth line, SSE/aN should be (SSE+SSTrts)/aN

Page 180 Example 5.7.1's first line should refer to Example 5.5.2 rather than 5.4.2

Page 184 Example 5.7.2 should be Example 5.7.4

Page 198 Line after (6.1.3), (autocovariance) not (autocorrelate)

Page 359 Display after (9.1.1), the second X\beta_1 should be X\beta_q

Page 366 (9.4.4) and the following display only need to hold for j greater than or equal to i

Page 366 Two lines below (9.5.2), "univariate model (4.1.2)" should be "univariate model (9.1.2)"

Page 374 Middle of page, ", we can replace \rho with \nu = P₀ \rho," is missing a transpose on P₀.

Page 375 In Exercise 9.4, T²=[n-r(X)]\sum. Multiply, not divide.

Page 375 In Exercise 9.4, subscripts on \phi's should be h not i

Page 376 Before penultimate display, reference to Exercise 9.3 should be to Exercise 9.2.

Page 377 Third display, the numerator of the approximation for theta should be [r(X)-r(X₀)] not [n-r(X)].

Page 378 Line 5, subscript on \phi should be h not i

Page 387 First line, the subscript on the identity matrix is 1 whereas it should be q

Page 395 Within Example 10.2.1, the reference to Example 12.2.2 should be to Example 12.6.1. (See correction on p. 475.)

Page 475 Example 12.2.2 should be labeled Example 12.6.1

Page 483 Example 12.6.1 should be labeled Example 12.7.1

Page 547 Line 1, "point" should be "pointed".

Page 576 In the second and third displays, the second derivative notation is pretty obvious but should have a double x subscript.

Page

Thanks to Penny Darsey for pointing out a large number of these.