# StatisticsBook

 Data Analysis in High Energy Physics - a Practical Guide to Statistical Methods Data Analysis in High Energy Physics - a Practical Guide to Statistical Methods, eds. Olaf Behnke, Kevin Kröninger, Thomas Schörner-Sadenius and Gregory Schott, Wiley-VCH, 2013. This practical guide covers the most essential statistics-related tasks and problems encountered in high-energy physics data analyses. It addresses both advanced students entering the field of particle physics as well as researchers looking for a reliable source on optimal separation of signal and background, determining signals or estimating upper limits, correcting the data for detector effects and evaluating systematic uncertainties. Each chapter is dedicated to a single topic and supplemented by a substantial number of both paper and computer exercises related to real experiments. A special feature of the book are the analysis walk-throughs used to illustrate the application of the methods discussed beforehand. The authors give examples of data analysis, referring to real problems in HEP, and display the different stages of data analysis in a descriptive manner.
 Contents Click on the chapter titles for more information like exercises, solutions, code etc. 1. Fundamental concepts (Roger Barlow) 2. Parameter estimation (Olaf Behnke and Lorenzo Moneta) 3. Hypothesis testing (Gregory Schott) 4. Interval estimation (Luc Demortier) 5. Classification (Helge Voss) 6. Unfolding (Volker Blobel) 7. Constrained fits (Benno List) 8. How to deal with systematic uncertainties (Rainer Wanke) 9. Theory uncertainties (Markus Diehl) 10. Statistical methods commonly used in high energy physics (Carsten Hensel und Kevin Kröninger) 11. Analysis walk-throughs (Aart Heijboer and Ivo van Vulpen) 12. Applications in astronomy (Harrison B. Prosper)
 Exercises and solutions A volume with all exercises and solutions will be made available soon. In addition, PDF files with the exercises and solutions for each individual chapter as well as the relevant code, macros, data files etc. will be made available in the box "Contents" above.
 Typos [page 7] Denominator of rhs of equation (1.16): "|V|" should be |V|^{1/2} [page 7] third line after eq. (1.18): The text should correctly read "of the mean is 68% obtained by calculating erf(y=1/sqrt(2))." [page 28] First line of "Efficiency" bullet item: "V" should be a matrix [page 28] Equation (2.3): Both "V" and "I" should be matrices [page 34] Equation (2.21): Both "V" and "I" should be matrices [page 70] First line of exercise 2.4: last time $t_n$ should be $t_N$. [page 71] Exercise 2.5: Argument x of function G(x) should be a vector. [page 71] Exercise 2.7: Times are given in minutes. [page 79] Example 3.1 (continued): the test size for t_c=4 is 0.043, not 0.046. [page 84] Equation (3.6): $\Phi$ is the cumulative distribution function ($\Phi^{-1}$ is the inverse). [page 85] Equation (3.8): must correctly read $Z = \sqrt{2 \ln Q}$ (the minus sign is wrong) [page87] Equation (3.14): Both $\nu_b$ and $\tau$ in both numerator and denominator should have double hats (not only a single hat) since none of the two fits of the likelihood function correspond to the global minimum. [page 87] Equation (3.14): The right-hand side is missing a $-2 \ln$. [page 321] Equation (9.22) and also five lines below: $g = v$ should be replaced by $F=v$. [page 379] Exercise 11.5 d): It should be $M_{\Zprime = \unit{251}{\GeV}$, not $M_{\Zprime = \unit{151}{\GeV}$.