`Modern Applied Statistics with S-Plus' by W.N. Venables & B.D. Ripley ====================================================================== Errata for the first printing of the second edition. That was written when 4.0 for Windows was in beta test, and some late changes were made. p. ii We supplied PostScript not LaTeX files. Dr Sheather's address is the Australian Graduate School of >Management<. p. 8 S-mode has been replaced by ESS, from http://franz.stat.wisc.edu/pub/ESS/ pp. 8,9 These instructions are not correct for S-PLUS 4.0. The main differences are (a) It is not necessary to make a _Data directory, but if you do, make a _Prefs directory as well. (b) Rather than setting the working or start in directory, add S_PROJ=path after the command/target. (c) Unless the project directory has both _Data and _Prefs directories, you will be asked if you want them created. Click on OK. (d) When S-PLUS 4.0 is launched, open a Commands Window from the button on the upper toolbar (the tool tips will help you to find which one this is). (If you always want a Commands Window at startup, select this from the menus via Options|General Settings...|Startup. This setting is saved on exiting the program.) p. 9 It is possible that steps 3 and 4 will not find the S-PLUS folder (although check if this has been given a different name). This will happen under NT4.0 when S-PLUS was installed under another account and so appears `below the line' on the Programs menu. In these cases you need use Find to locate the program splus.exe and make a shortcut to that in the Programs folder (or create a new folder within that folder and put the shortcut there). p. 10 It is not necessary to use library(helpfix, first=T) with S-PLUS 4.0 for Windows. p. 43 There is a new function regexpr() in 4.0 for Windows that behaves like grep() on Unix. It is not used in objects(), though. p. 45 The Object Manager in 3.x for Windows is replaced by the object browser, invoked from the toolbar. p. 49 l.-15 at least >1<. Arrays can be of dimension 1 (although this is not very useful). p. 56 The footnote is no longer needed. Note that 4.0 has a similar function ginverse. p. 67 l.-11 The release version of 4.0 echoes prompts and error messages to standard output. p. 67 l.-8 In S-PLUS 4.0 the BATCH command is Splus [S_PROJ=dir] /BATCH infile [[outfile] errfile] and errors (but not prompts) are written to the outfile if errfile is not present. p. 68 l.-5 median |x - \mu| (| | not [ ]). p. 71 Table 3.2. S-PLUS 4.0 introduced a device pdf.graph to produce output in PDF (Adobe's Portable Document Format, sometimes known as Acrobat format). p. 72 printgraph only exists in Unix versions. p.112 l.13 use table and >pie or< piechart. p.119 l.-2 sweep(A, 2, apply(A,2,mean), "-") p.136f print methods should have first argument x, so change object to x throughout print.sum.my.t.test. p.150 l.-3 invisible(y) (not x), as print method should return their first argument unchanged. p.151, 3 The incorrect versions of Integrate with a scalar return value give an error in S-PLUS 4.0 and later. p.168 l.-3 estimate of the >skewness<. p.179 (5.4) 1/b should be 1/(nb) p.183 l.3 summation is over i, not j. p.215 The calculation of Vars fails in S-PLUS 4.0, since var reports an error (rather than NA) for less than 2 cases. p.277 first display. theta/log_2 p.292 To work around a bug in deriv() introduced in S-PLUS 4.0, mix.gr needs to be altered to mix.gr <- function(p, x) { u1 <- p[2]; s1 <- p[3]; u2 <- p[4]; s2 <- p[5]; p <- p[1] e <- lmix2a(x, p, u1, s1, u2, s2) rep(1, length(x)) %*% attr(e, "gradient") } p.313 The correlation here is in fact between observations, as lme has taken the ranks of the times. Set serial.covariate.transformation = "none" to achieve what we intended (but note this fails to fit correctly). p.327 display. ... - \alpha . (This is necessary as the smoothers used by gam() do not include the constant.) p.340 l.7,8 x is the independent variables, y the dependent ones. p.346 (12.3) The S code follows Fleming and Harrington in breaking ties, so if there were 3 deaths from 10 at risk the summand would be 1/10 + 1/9 + 1/8. p.347 third displayed equation. - is needed before H(t). p.354 Figure 12.3 is of the >leuk< dataset p.362 l.17 The scaling used in the Schoenfeld residuals is actually to give I^{-1}schoenfeld where I is the (partial) information matrix from the Cox model. p.370 We should have had age=c(50,50) - 48, but in fact survfit would in error here even with the correct age. p.368 l.-2 `interval-censored' means different things to different people: here it means both left- and right-censored. p.397 l.18 The first term of (13.3) is the >squared< Mahalanobis ... p.400 There is a misprint in the table of Fisher's data: (medium, black) had a count of 26 not 25. The corrected results are First canonical correlation: 0.44637 Row scores: blue light medium dark -0.89679 -0.98732 0.075306 1.5743 Column scores: fair red medium dark black -1.2187 -0.52258 -0.094147 1.3189 2.4518 p.411 (13.9) The scond Psi should be inverted too. p.416 Caption: The numbers m/n denote the proportion of training cases reaching that node that are classified incorrectly by the label in the node p.431 rts(scan(n=48)), not 47. p.441 l.2 sqrt(m) + 0.11 + 0.12/sqrt(m). p.457 The coh component is the _squared_ coherency, so these plots need some adjustment (and people need to use more helpful names). p.478 The 3D spherical covariance should have r/d rather than d/r, twice. p.487 l.6 >log< p(c | x) p.489 l.-10 For >classification trees< we can use p.491 l.-4 Figure 17.3, not 17.4. p.495 With the current version of predict.tree in library treefix use function(obj, x) predict(obj, fgl[x, ], type="class") instead of function(obj, x) levels(fgl$type)[predict(obj, fgl[x, ], type="class")] For additional information on neural networks and LVQ see the on-line complements (as well as the script). p.530 Ripley (1997) page numbers are 110-6. p.533ff The front matter was changed by Springer, so change all the index references of v to vii and of xiv to xvii. p.542 offset. Add page 229 as the primary reference. Typos ----- p. 17 l.-3 >A< map p. 41 l.1 so >the< first p. 91 l.-3 The axes >are< omitted p. 98 footnote pscale>s< p.154 l.6 >A< small example p.167 first line after code. In S-PLUS 3.2 this >has< been modified.. p.169 in nclass.Scott, nclass.FD, ^(-1/3) would be better. p.170 l.3 it is easy >to< write one's p.177 l.-1 n^{-1/5} p.179 ll.-19,-18 "rectangular", "triangular" and "cosine" should be in typewriter font. p.183 l.11 using the distances >between< all pairs of points p.238 l.-7 For large models >it< is often helpful to p.253 l.-8 with one value >that< appears to be out by p.327 l. 22 to fit >locally< a polynomial p.331 lines -4,-3 The suffix on alpha should be j not k, twice. p.341 l. -7 Friedman's (1991) >MARS< explictly ... pp.359, 529 Nagelkerke p.388 l. 16 Add comma after Gabriel, 1971. p.454 l. -1 missing X after \Delta \Delta_{12} p.455 l. 3 missing X after \Delta \Delta_{12} p.456 l.-5 missing ( after I_ii p.490 l. 2 >logistic< regression p.498 l. 2 ... layer >and< for p.522 Box & Cox >An< analysis ...