---

Useful Information about the midterm exam

• The exam covers all material:
  • In Chapters Two and Three of your textbook
  • Covered in class up to October 9
    
    
• The exam is:
  • Closed book
  • Closed notes
    
    
• You should bring to class a calculator
  
  
• You should bring to class up to twenty pages of personal notes. You can turn these notes in at the end of the midterm exam if you wish. I will judge these notes for organization and clarity though they can be hand written. I will give a grade to these notes and used this grade in place of the grade of your worse exam question provided the grade of the notes is better.
  
  
• Your notes should contain information that includes but is not limited to the following:
  • Formulas for the distribution, failure rate, and density of each of the following models
    • Exponential
    • Weibull
    • Gumbel
    • Lognormal
    • Generalized Gamma
      • Relationship to the models above
  • Interpretation of the parameters of the models above, e.g.,
    • What in one interpretation of the Weibull scale parameter?
    • What information about the hazard rates in contained in the Weibull shape parameter?
    • Relationships among the parameters of the Weibull for the data and Gumbel for the log data.
  • Origins of other models such as the Gamma model
  • Formulas for method of moments estimators
  • Formulas used in conjunction with the empirical survivor function and the Product-Limit estimator
    • Survivor function estimator
    • Plotting points
    • Standard errors and confidence limits for the survivor function
    • Standard errors and confidence limits for the integrated hazard
  • Formulas used to draw QQ plots and to obtain graphical estimates of model parameters
    • Exponential
    • Weibull
    • Gumbel
    • Lognormal
  • Interpretations of the slope of the empirical integrated hazard (exponential QQ) plot
  • Competing risks
    • Model
    • Assumptions
    • Interpretations
    • Practical significance
  • Formulas used in conjunction with maximum likelihood estimation
    • Information matrix
    • Variance-covariance matrix
  • Formulas for the MLEs and standard errors in the exponential model
    • Hazard rate
    • Mean time to failure
  • Delta Method formulas
    • Univariate
    • Bivariate
    • Trivariate
  • Particular functions of the parameters used in conjunction with the Delta Method to calculate standard errors of quantiles and survival probabilities for each of the following models
    • Exponential
    • Weibull
    • Gumbel
    • Lognormal
  • Formulas for Likelihood Ratio test and confidence regions
    • How to determine the degrees of freedom
    • How to interpret the distance between maximized log-likelihoods in nested models
  • Formulas for tests and confidence regions based on the variance-covariance matrix
    • Comparison to likelihood ratio methods
  • Formulas used to draw PP and SP plots
    • Exponential
    • Weibull
    • Gumbel
    • Lognormal
  • A standard normal table in case you have to calculate the standard error of a lognormal quantiles
  • A Chi-Square table of critical values in case you are asked to calculate a likelihood ratio tests or another maximized likelihood procedure needing these critical values
  • Table for interpreting the result of the Weibull versus lognormal tests given in the notes
    
    
• Other recommendations
  • Know how to interpret all JMP output that you have seen in class
  • Know how to interpret all graphs and tables in the book
  • Review the study questions at the end of the class summaries provided in this course homepage
  • Be prepared to discuss the relative merits of the different procedures discussed in class.

Take a look at some of my old exams

Do you have something to tell me?