- Paperback: 500 pages
- Publisher: Xlibris, Corp. (September 13, 2010)
- Language: English
- ISBN-10: 1453519106
- ISBN-13: 978-1453519103
- Product Dimensions: 6 x 1.1 x 9 inches
- Shipping Weight: 1.8 pounds (View shipping rates and policies)
- Average Customer Review: 3 customer reviews
- Amazon Best Sellers Rank: #4,920,588 in Books (See Top 100 in Books)
Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
Signal From Noise, Information From Data
Use the Amazon App to scan ISBNs and compare prices.
Showing 1-3 of 3 reviews
There was a problem filtering reviews right now. Please try again later.
"We calculate [...] by summing the reciprocals of each observation [...] and dividing that sum into the number of observations." (pg. 77)
The formula that follows is not N divided by the sum of reciprocals but is instead N divided by the reciprocal of the sum, which is something different. On pg. 84 the correct formula appears, and on the same page another series of errors connected to their "explanation" of the harmonic mean:
"Rearranging the equation [...] makes it obvious that the harmonic mean is the mean of the reciprocals..."
The equation I removed from the quote is ALSO NOT the mean of the reciprocals, but is instead the sum of the reciprocals divided by 1/N, and that is AGAIN something different. In the next line a faulty calculation is included which AGAIN makes the error of equating the sum of reciprocals with the reciprocal of a sum.
Do the authors not understand basic arithmetic? It's difficult to comprehend how the authors of a mathematical topic could be so sloppy both in explanation and in calculation, and because of this I'm weary of continuing through the text. How can I trust what I'm reading? Other reviewers have given high-star ratings on the basis of the text's "gentleness", but this is surely eclipsed by the sheer number of explicit errors. The example above is just one of many bumbled topics presented in the text. Very embarrassing. Aside from the very obvious errors in the text, the exercises are outdated and quite useless: am I really expected to write out, by hand no less, calculations concerning 5 variables each having 50 values, a total of 250 numbers (chapter 2 exercises)? In the real world working with data means computationally interacting with datasets using software and code. One expects a much higher degree of fidelity with modern techniques from a 2010 text. Again, how did this happen?