True or False

This is something that I keep tripping up on in R programming. True is all caps when used as a logic operator. Same with false. Type TRUE when I want to know if something is true or set it to true. Same for false, type FALSE. And don’t leave the caps lock on.

Posted in Uncategorized | Tagged ,

The Cycle of Deviant Behavior


The Cycle of Deviant Behavior by Howard B. Kaplan and Glen C. Tolle,Jr. published by springer press. 2006

I came across this book on a pertinent topic. The book is an excellent example on how to conduct research studies. It is a longitudinal prospective study. The gold standard of this type of research study. It is not retrospective like so many studies are. The book has a thorough literature review. Well documented handling of missing values. Uses structural equation modeling. The book provides a template for systematic logic of procedures for analysis of bi variate relationships.

Posted in Uncategorized | Tagged

Measuring Data Quality for Ongoing Improvement


Measuring Data Quality for Ongoing Improvement by Laura Sebastian-Coleman.

This the kind of book I like finding and recommending. DQAF, Data Quality Measurement Framework. The book brings quality control to data management. The first section of the book thoroughly covers definitions. It then goes on to explain measure. Continuing on to the how part of data quality. The book is is a useful reference book. I know that I will use it as a reference for upcoming writing and talks.

Posted in Uncategorized | Tagged , ,

Bright Shiny New Year 2014

Actually it is cold grey and foggy. I am forever an optimist. We had winter solstice and the days are slowly getting longer. I get distracted by bright shiny things. Bright shiny things that attracted my attention presently are: D3, amazing things with static data. Text mining using R. Maps, geo.json, github renders maps. Tensors, I am working on how to explain them better. Now to get writing.

Posted in Uncategorized

Handbook of Partial Least Squares, Concepts, Methods and Applications



Handbook of Partial Least Squares, Concepts, Methods and Applications. Edited by V. Esposito Vinzi, W.W. Chin, J Hensley and H. Wang. Published by Springer.

This Handbook is a book of  33 papers selected from three rounds of peer review process. There is a lot of very good material in this book. I wish that I had delved into it earlier

Chapter 28 which is a paper on How to Write Up and Report PLS Partial Least Squares Analyses, discusses Sample size and goes into detail about how and why you can use a smaller sample size with PLS. That alone is enough reason to read this book. Add to it the tables and examples there is enough material to keep me really busy reading the next time I need to do a questionnaire.

Posted in Uncategorized | Tagged

Smoothing Spline ANOVA Models



Smoothing Spline ANOVA Models by Chong Gu, published by Springer

I am intrigued by this book. Splines were my favorite thing in graduate school. I have made a lot of ANOVA models. It is fun to see what I tried to do finally achieved.
Smoothing Splines ANOVA Models uses R as the programming language. Great to see in a book are in depth proofs and R code.
Chapter 3.3 shows how to draw Bayesian confidence intervals in R.

Chapter 3.10.1 discusses the difference between natural splines and B splines. That B splines have different boundary conditions.
There is code for doing cubic splines with a jump. Something that you run into with real data.
In Chapter 8.63 about hazard functions and the Weibull family has code for cubic spline Weibull regression with censored and truncated data.
I am enjoying reading this book. The code works and the examples are easy to understand.

Posted in Uncategorized | Tagged , ,

Linear Mixed-effects Models Using R


Linear Mixed-effects Models Using R by Andrzej Galecki and Tomasz Burzkowski, published by Springer is a book that covers in dept a lot of material on linear models.

The book has clear instructions on how to program in R.

The book in chapter 4 covers model reduction using a null model and alternative model, which are nested models. Model reduction is a topic that needs to be discussed by coders. I have talked with many people who have put everything in a regression model just because they could.
Section 5.2 has the proper form for model formulas
R expression ~ term.1 + term.2 + …+ term.k
It is nice to see this spelled out so clearly.

Chapter 8 shows how to use the nlme package.

Part Three covers Lm’s that allow the relaxing of the assumptions of independence and variance of homogeneity. This a topic that I needed information on.

This is a good reference book.

Posted in Uncategorized | Tagged