gjkerns / ipsur

IPSUR, Introduction to Probability and Statistics Using R, is a textbook written for an undergraduate course in probability and statistics. IPSUR is FREE, in the GNU sense of the word.

IPSUR: Introduction to Probability and Statistics Using R

Copyright (C) 2022 G. Jay Kerns

Version 3.0

This package is the source code to generate IPSUR, a free textbook on Probability and Statistics.

Install

This package is not distributed on CRAN. The easiest/quickest way to get IPSUR (which at the moment doesn't do what it is supposed to do) is to install the devtools package if you don't have it already and run

devtools::install_github("gjkerns/prob")
devtools::install_github("gjkerns/IPSUR")

You will get lots of errs due to missing dependencies. See the file renv.lock to understand which packages you need to run IPSUR---there are quite a few.

The source code for IPSUR is released under the GNU General Public License, version 3. See the file COPYING for details.

Load

Once IPSUR is downloaded and installed, it must be loaded into R. Load the package with the library function:

library(IPSUR)

and finally launch your PDF viewer with

read(IPSUR)

ipsur's Issues

need to fix bivariate normal plot; rotate -90 degrees

Change stemplot to stem-and-leaf display throughout

This matches John Tukey's language in Exploratory Data Analysis (1977).

Remark 5.1

Remark 5.1. Note that although we say X is 3.5 on the average, we must keep in mind that our X
never actually equals 3.5 (in fact, it is impossible for X to equal 3.5).

should be

Remark 5.1. Note that although we say X is 1.5 on the average, we must keep in mind that our X
never actually equals 1.5 (in fact, it is impossible for X to equal 1.5).

add back library() calls where needed

add savehistory to Intro to R

title page for Second Edition

relevel function to reset the reference level

Add drawbacks of MAD from paper in PFDS

Add link to R-Bloggers in External Resources. And CRANtastic and ...

go through and check all figure dimensions

Double-check the confidence interval example in Chapter 9.

Make sure the random plot matches the language in the narrative.

Equation labels are not working right. Fix them.

fig-philosophy not in output.

Figure 11.1.1 (Philosophical foundations of SLR) did not come out and leads to a broken link in section 11.1.1. pg. 227 (IPSUR, 2nd edition June 6, 2014)

The relevant code is in IPSUR.org lines 13543. Unfortunately, I do not know org-mode so I do not know where the mistake may be (the R code works)

https://github.com/gjkerns/IPSUR/blob/master/IPSUR.org#L13643

bptest should probably studentized = FALSE

have students check assumptions for testing/CI's without plug-chug formulas

add exercises

aov and anova, read ?aov and ?anova, want balanced designs, don't forget replications function

stripplot for lattice

mention about higher moments needing huge sample size and being sensitive to outliers, and L-moments

Move discussion of sampling distr of xbar - ybar from confidence intervals to sampling distributions. Talk about the pooled estimator.

also check out ?cld in package multcomp

add descriptive code at beginning of each chapter for tangle

standard deviation is not a good measure of spread for asymmetric distributions.

Use the median and quartiles instead.

Example 4.34 or Section 4.6.4

Original text:

What is the probability of observing "red", then "green", then "red"?

prob(N, isin(N, c("red", "green", "red"), ordered = TRUE))
[1] 0.175
Note that the exact value is 7/20 (do it with the Multiplication Rule).

Should say:

What is the probability of observing "red", then "green", then "red"?

prob(N, isin(N, c("red", "green", "red"), ordered = TRUE))
[1] 0.175
Note that the exact value is 7/40 (do it with the Multiplication Rule).

Justification: (7/10)(3/9)(6/8)=126/720=7/40=.175

need to update bibliography

change clear screen to Ctrl+L for emacs+ubuntu

To transform a factor f to approximately its original numeric values, as.numeric(levels(f))[f] is recommended

fix order of max likelihood graphs fish examples

Example 4.26

Example 4.26 says:

(11 * 10 * 9) * (7 * 6 * 5 * 4) * 313
[1] 260290800
or alternatively
prod(9:11) * prod(4:7) * 313
[1] 260290800
or even
prod(factorial(c(11, 7))/factorial(c(8, 3))) * 313
[1] 260290800

It should say:

(11 * 10 * 9) * (7 * 6 * 5 * 4) * 31^3
[1] 24774195600
or alternatively
prod(9:11) * prod(4:7) * 31^3
[1] 24774195600
or even
prod(factorial(c(11, 7))/factorial(c(8, 3))) * 31^3
[1] 24774195600

need to do index

xyplot(eruptions ~ waiting, data = faithful, type =c("p","r")) will plot a regression line on the plot

Installing from github

Hello,
I noticed that IPSUR is noticeably more recent on github than on the R repository. I would like to use the newer one but I'm a pure beginner and unfortunately don't know how to install it from github. Could you please write a short install explanation ?

thanks a lot

add quotes to all library() calls, and if I am going to do this, then I might as well delete read("IPSUR")

make sure to talk about regression through the origin and how it screws up the p-value