[Distr-commits] r601 - in branches/distr-2.2/pkg/distr: inst/doc man
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Tue Oct 6 14:16:11 CEST 2009
Author: ruckdeschel
Date: 2009-10-06 14:16:11 +0200 (Tue, 06 Oct 2009)
New Revision: 601
Modified:
branches/distr-2.2/pkg/distr/inst/doc/Rplots.pdf
branches/distr-2.2/pkg/distr/man/qqbounds.Rd
Log:
distr: qqplot: some minor typos in qqbounds.Rd
Modified: branches/distr-2.2/pkg/distr/inst/doc/Rplots.pdf
===================================================================
--- branches/distr-2.2/pkg/distr/inst/doc/Rplots.pdf 2009-10-06 11:52:29 UTC (rev 600)
+++ branches/distr-2.2/pkg/distr/inst/doc/Rplots.pdf 2009-10-06 12:16:11 UTC (rev 601)
@@ -2,8 +2,8 @@
%âãÏÓ\r
1 0 obj
<<
-/CreationDate (D:20091006135110)
-/ModDate (D:20091006135110)
+/CreationDate (D:20091006141539)
+/ModDate (D:20091006141539)
/Title (R Graphics Output)
/Producer (R 2.10.0)
/Creator (R)
@@ -47,7 +47,7 @@
38.97 172.78 m 148.04 172.78 l S
38.97 227.22 m 148.04 227.22 l S
38.97 281.66 m 148.04 281.66 l S
-38.97 336.10 m 148.04 336.10 l S
+38.97 336.09 m 148.04 336.09 l S
38.97 390.53 m 148.04 390.53 l S
38.97 444.97 m 148.04 444.97 l S
0.000 0.000 0.000 RG
@@ -275,14 +275,14 @@
65.15 74.45 l
65.25 74.75 l
65.35 75.05 l
-65.45 75.37 l
+65.45 75.36 l
65.55 75.68 l
65.65 76.01 l
65.75 76.34 l
65.85 76.68 l
65.96 77.03 l
66.06 77.39 l
-66.16 77.76 l
+66.16 77.75 l
66.26 78.13 l
66.36 78.51 l
66.46 78.90 l
@@ -315,7 +315,7 @@
69.19 93.26 l
69.29 93.95 l
69.39 94.66 l
-69.49 95.37 l
+69.49 95.38 l
69.59 96.11 l
69.70 96.85 l
69.80 97.61 l
@@ -367,7 +367,7 @@
74.45 149.49 l
74.55 151.02 l
74.65 152.57 l
-74.75 154.14 l
+74.75 154.13 l
74.85 155.72 l
74.95 157.32 l
75.05 158.94 l
@@ -375,12 +375,12 @@
75.26 162.23 l
75.36 163.91 l
75.46 165.60 l
-75.56 167.30 l
+75.56 167.31 l
75.66 169.03 l
75.76 170.77 l
75.86 172.53 l
75.96 174.31 l
-76.06 176.11 l
+76.06 176.10 l
76.17 177.92 l
76.27 179.74 l
76.37 181.59 l
@@ -416,7 +416,7 @@
79.40 243.87 l
79.50 246.13 l
79.60 248.41 l
-79.70 250.69 l
+79.70 250.70 l
79.81 252.99 l
79.91 255.29 l
80.01 257.60 l
@@ -424,14 +424,14 @@
80.21 262.24 l
80.31 264.57 l
80.41 266.90 l
-80.51 269.25 l
+80.51 269.24 l
80.61 271.59 l
80.72 273.94 l
80.82 276.30 l
80.92 278.66 l
81.02 281.02 l
81.12 283.39 l
-81.22 285.75 l
+81.22 285.76 l
81.32 288.13 l
81.42 290.50 l
81.52 292.87 l
@@ -466,7 +466,7 @@
84.46 359.93 l
84.56 362.10 l
84.66 364.26 l
-84.76 366.39 l
+84.76 366.40 l
84.86 368.52 l
84.96 370.63 l
85.06 372.71 l
@@ -493,7 +493,7 @@
87.19 411.66 l
87.29 413.24 l
87.39 414.79 l
-87.49 416.31 l
+87.49 416.32 l
87.59 417.81 l
87.69 419.27 l
87.79 420.70 l
@@ -583,7 +583,7 @@
96.28 413.39 l
96.39 411.84 l
96.49 410.26 l
-96.59 408.65 l
+96.59 408.66 l
96.69 407.02 l
96.79 405.37 l
96.89 403.69 l
@@ -606,7 +606,7 @@
98.61 371.75 l
98.71 369.70 l
98.81 367.63 l
-98.91 365.55 l
+98.91 365.56 l
99.01 363.46 l
99.11 361.36 l
99.22 359.24 l
@@ -623,7 +623,7 @@
100.33 335.17 l
100.43 332.93 l
100.53 330.68 l
-100.63 328.43 l
+100.63 328.42 l
100.73 326.16 l
100.83 323.90 l
100.93 321.63 l
@@ -631,12 +631,12 @@
101.14 317.07 l
101.24 314.79 l
101.34 312.50 l
-101.44 310.21 l
+101.44 310.22 l
101.54 307.92 l
101.64 305.63 l
101.74 303.34 l
101.84 301.05 l
-101.95 298.75 l
+101.95 298.76 l
102.05 296.46 l
102.15 294.17 l
102.25 291.88 l
@@ -657,19 +657,19 @@
103.77 257.93 l
103.87 255.72 l
103.97 253.51 l
-104.07 251.31 l
+104.07 251.30 l
104.17 249.11 l
104.27 246.93 l
104.37 244.75 l
104.47 242.59 l
104.57 240.43 l
104.68 238.28 l
-104.78 236.15 l
+104.78 236.14 l
104.88 234.02 l
104.98 231.90 l
105.08 229.80 l
105.18 227.71 l
-105.28 225.63 l
+105.28 225.62 l
105.38 223.56 l
105.48 221.50 l
105.59 219.45 l
@@ -682,9 +682,9 @@
106.29 205.49 l
106.39 203.55 l
106.49 201.63 l
-106.60 199.72 l
+106.60 199.71 l
106.70 197.82 l
-106.80 195.94 l
+106.80 195.93 l
106.90 194.07 l
107.00 192.21 l
107.10 190.37 l
@@ -703,7 +703,7 @@
108.42 167.87 l
108.52 166.26 l
108.62 164.65 l
-108.72 163.07 l
+108.72 163.06 l
108.82 161.49 l
108.92 159.94 l
109.02 158.40 l
@@ -724,7 +724,7 @@
110.54 137.28 l
110.64 136.00 l
110.74 134.74 l
-110.84 133.49 l
+110.84 133.50 l
110.94 132.26 l
111.04 131.05 l
111.15 129.85 l
@@ -861,7 +861,7 @@
124.39 66.88 l
124.49 66.80 l
124.59 66.72 l
-124.69 66.63 l
+124.69 66.64 l
124.79 66.56 l
124.89 66.48 l
125.00 66.40 l
@@ -1093,7 +1093,7 @@
38.97 172.78 m 34.21 172.78 l S
38.97 227.22 m 34.21 227.22 l S
38.97 281.66 m 34.21 281.66 l S
-38.97 336.10 m 34.21 336.10 l S
+38.97 336.09 m 34.21 336.09 l S
38.97 390.53 m 34.21 390.53 l S
38.97 444.97 m 34.21 444.97 l S
BT
Modified: branches/distr-2.2/pkg/distr/man/qqbounds.Rd
===================================================================
--- branches/distr-2.2/pkg/distr/man/qqbounds.Rd 2009-10-06 11:52:29 UTC (rev 600)
+++ branches/distr-2.2/pkg/distr/man/qqbounds.Rd 2009-10-06 12:16:11 UTC (rev 601)
@@ -34,38 +34,50 @@
For simultaneous intervals,
the finite sample version is based on C function \code{"pkolmogorov2x"}
from package \pkg{stats}, while the asymptotic one uses
- R function \code{pkstwo} from package \pkg{stats}. Both use the fact,
+ R function \code{pkstwo} again from package \pkg{stats}, both taken
+ from the code to \code{\link[stats:ks.test]{ks.test}}.
+
+ Both finite sample and asymptotic versions use the fact,
that the distribution of the supremal distance between the
empirical distribution \eqn{\hat F_n}{F.emp} and the corresponding theoretical one
\eqn{F} (assuming data from \eqn{F})
does not depend on \eqn{F} for continuous distribution \eqn{F}
- and leads to the Kolmogorov distribution. In case of \eqn{F} with jumps the
- corresponding Kolmogorov distribution is used to produce conservative intervals.
+ and leads to the Kolmogorov distribution (compare, e.g. Durbin(1973)).
+ In case of \eqn{F} with jumps, the corresponding Kolmogorov distribution
+ is used to produce conservative intervals.
\cr
For pointwise intervals,
- the finite sample version is based corresponding binomial distributions,
- while the asymptotic one uses a CLT approximation (in fact only valid for
- distributions with strictly positive density at the evaluation quantiles).
+ the finite sample version is based on corresponding binomial distributions,
+ (compare e.g., Fisz(1963)), while the asymptotic one uses a CLT approximation
+ for this binomial distribution. In fact, this approximation is only valid
+ for distributions with strictly positive density at the evaluation quantiles.
+
In the finite sample version, the binomial distributions will in general not
be symmetric, so that, by setting \code{nosym.pCI} to \code{TRUE} we may
produce shortest asymmetric confidence intervals (albeit with a considerable
- computational effort). The symmetric intervals returned by default will
+ computational effort).
+
+ The symmetric intervals returned by default will
be conservative (which also applies to distributions with jumps in this case).
+
For distributions with jumps or with density (nearly) equal to 0 at the
corresponding quantile, we use the approximation of \code{(D-E(D))/sd(D)}
by the standard normal at these points; this latter approximation is only
- available if \code{require(distrEx)==TRUE}; otherwise the corresponding
+ available if package \pkg{distrEx} is installed; otherwise the corresponding
columns will be filled with \code{NA}.
}
\value{
A list with components \code{crit} --- a matrix with the lower and upper confidence
bounds, and \code{err} a logical vector of length 2.
- Components \code{crit} is a matrix with 4 columns and \code{length(x)} rows
- with columns \code{c("sim.left","sim.right","pw.left","pw.right")}.
+
+ Component \code{crit} is a matrix with \code{length(x)} rows
+ and four columns \code{c("sim.left","sim.right","pw.left","pw.right")}.
Entries will be set to \code{NA} if the corresponding \code{x} component
is not in \code{support(D)} or if the computation method returned an error
or if the corresponding parts have not been required (if \code{withConf.pw}
- or \code{withConf.sim} is \code{FALSE}). \code{err} has components \code{pw}
+ or \code{withConf.sim} is \code{FALSE}).
+
+ \code{err} has components \code{pw}
---do we have a non-error return value for the computation of pointwise CI's
(\code{FALSE} if \code{withConf.pw} is \code{FALSE})--- and \code{sim}
---do we have a non-error return value for the computation of simultaneous CI's
@@ -75,6 +87,7 @@
Durbin, J. (1973)
\emph{Distribution theory for tests based on the sample distribution
function}. SIAM.
+
Fisz, M. (1963). \emph{Probability Theory and Mathematical Statistics}.
3rd ed. Wiley, New York.
}
More information about the Distr-commits
mailing list