Take the 2-minute tour ×
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required.

I am writing a dissertation in which I use maximum likelihood estimation and an alternative Bayesian approach. I have written up the maximum likelihood estimation approach. However, I need some advice on presenting the Bayesian part. Can someone suggest me a source which can shape my thoughts so that I can write up the Bayesian part. I need to present the fitted values of a linear model and a logit model and also their Bayesian intervals via plots as the sample size is too large to be given in the dissertation. I want to include winBUGS plots. What plots can be useful to give the fitted values and their Bayesian intervals.

Any advice would be greatly appreciated.

share|improve this question
    
Is your linear model a classical linear regression model ($Y=X\beta+\epsilon$) ? And your logit model, is it the logistic regression model ? –  Stéphane Laurent Jan 10 '13 at 18:34
    
My linear model is one-way error components random effects model whose specification is y_i = 1 - xbeta + delta_i + e_i where x is the design matrix, beta is the vector of the regression coefficients, delta_i is the i-th random effect, e is the i-th usual error and yes, my logit model is the logistic regresssion model –  Günal Jan 10 '13 at 18:44
    
Ok. You have written "the sample size is too large to be given in the dissertation", what did you mean ? –  Stéphane Laurent Jan 10 '13 at 19:14
    
I meant that I have 2456 fitted values in the linear model and 1002 in the logit model and also their bayesian intervals. So, I cannot give these numbers directly. Therefore, I need to plot these fitted values and bayesian intervals. Also, I want to the reader to distinctly see that I have used Winbıugs –  Günal Jan 10 '13 at 20:39
    
Can anyone suggest any source regarding my query please? –  Günal Jan 11 '13 at 19:10

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.