Solved – Book for broad and conceptual overview of statistical methods

machine learningrreferencesregressionsimulation

I am very interested about the potential of statistical analysis for simulation/forecasting/function estimation, etc.

However, I don't know much about it and my mathematical knowledge is still quite limited — I am a junior undergraduate student in software engineering.

I am looking for a book that would get me started on certain things which I keep reading about: linear regression and other kinds of regression, bayesian methods, monte carlo methods, machine learning, etc.
I also want to get started with R so if there was a book that combined both, that would be awesome.

Preferably, I would like the book to explain things conceptually and not in too much technical details — I would like statistics to be very intuitive to me, because I understand there are very many risky pitfalls in statistics.

I am off course willing to read more books to improve my understanding of topics which I deem valuable.

Best Answer

  • Maybe you'd like something like Data Analysis and Graphics Using R: An Example-Based Approach by John Maindonald and W. John Braun

    • Website for book
    • Amazon with assorted reviews
    • I recommend it because the book ticks a few of your boxes; it teaches a little R; it provides an overview of a range of different modelling techniques (e.g., multiple regression, time series, graphics, generalised linear model, etc.) without going into too much mathematical detail; it's fairly applied.

  • I agree with @Greg Snow that you may be better off thinking in terms of reading a number of different books. For each topic you mentioned (e.g., Bayesian statistics, time series, simulations, R, machine learning) there are good books dedicated to that particular topic. You may wish to ask separate questions about what would be a good book given your particular interests in that topic.

  • Good freely available online options

    • Elements of Statistical Learning is an excellent book and is even available online for free. From your post, I get the sense that it might be a little more technical than you want at first, but check it out and see what you think. Maybe you'll be ready for it now; maybe later.
    • Benjamin Bolker's Ecological Models and Data in R is another good one. It is from an ecology perspective, but does explain simulation and model fitting clearly from a relatively non-technical perspective; and it's all implemented in R. You can see all his R code on the website. You can even see the Sweave documents used to generate the book!
    • There's a good list of free R documentation on CRAN with some of the documents also providing broader instruction on statistics.
Related Question