Course Meeting Times

Lectures: 3 sessions / week, 1 hour / session


This course is a broad treatment of statistics, concentrating on specific statistical techniques used in science and industry. Topics include: hypothesis testing and estimation, confidence intervals, chi-square tests, nonparametric statistics, analysis of variance, regression, correlation, decision theory, and Bayesian statistics.


Probability and Random Variables (18.440) or Probabilistic Systems Analysis (6.041) or the equivalent (one semester of calculus-based probability). The prerequisite material is also covered in the required textbook (see below), Chapters 1-5.


Buy at Amazon Rice, John A. Mathematical Statistics and Data Analysis. 3rd ed. Belmont, CA: Duxbury Press, 2006. ISBN: 9780534399429.

We will cover material from Chapters 6 to 15 of the book. Some supplements will be made available as we go along. We will review probability as needed.

Versions of 18.443 Taught at MIT

From fall 2003 through spring 2009, 18.443 has been taught in different flavors in the fall and spring. In the fall, it has been relatively more parametric or indeed Bayesian (a probability distribution being assumed a priori on the unknown parameters). The Bayesian textbook by DeGroot and Schervish (Buy at Amazon DeGroot, Morris H., and Mark J. Schervish. Probability and Statistics. 3rd ed. Reading, MA: Pearson Addison Wesley. ISBN: 9780201524888.) has been assigned although not necessarily closely followed.

In the spring, the textbook by J. Rice (see above), which is relatively less parametric and has only a small amount of Bayesian material, has been assigned. Some supplementary Bayesian material has been included in the course.

A special feature of the Fall 2006 version by Professor Panchenko was the use of Mathematica.


There will be 10 problem sets. None will be due in weeks when there are exams.


There will not be a final exam. During the term there will be three exams, equally spaced during the course.


Three exams (25% each) 75%
Problem sets 25%