Economic and Official statistics

There are two groups of problems and a paper to read and comment on. You need to chose one problem from each group, and you need to choose one of the papers below, read it and write a short (short means at most two pages) opinion on it. If a problem requires calculations or simulations you can use the statistical package of your choice. Include only the essential graphs and/or tables in your solutions. Do not spend time editing the solutions. Use the time to read good statistical books instead. There is no deadline but you get the grade once the seminar assignment is completed.

Group 1:

• Cluster sampling estimation 1
• Cluster sampling estimation 2
• Estimating net asset value
• Randomized response in sampling
• Sample size in clinical trials
• Parameter estimation for Weibull distribution
• Parameter estimation for the generalized gamma distribution
• Parameter estimation for autoregressive time series

Group 2:

• Wilks' theorem and contingency tables
• Analysis of variance
• Testing for subspace for multivariate normal distribution
• Correlation test for multivariate normal distribution
• Regression with dependent errors
• Logistic regression for kyphosis (data)
• Probit model (data)
• Logit model (data)
• Generalized regression

Papers:

• L. Breiman, Statistical Modeling: The Two Cultures, Statistical Science, 16(3), 2001, 199-231.
• S. T Ziliak, D. N. McCloskey, The Cult of Statistical Significance, Joint Statistical Meetings 2009, Section of Education.
• W. Krämer, The Cult of Statistical Significance, What economists should and should not do to make their data talk, Working Paper Series of the German Data Forum, 2011, Federal Ministry of Edication and Research.
• H. V. Roberts, Application in Business and Economic Statistics: Some Personal Views, Statistical Science, 5(4), 1990, 372-402.