Forskjeller

Her vises forskjeller mellom den valgte versjonen og den nåværende versjonen av dokumentet.

Lenk til denne sammenligningen

Begge sider forrige revisjon Forrige revisjon
Neste revisjon
Forrige revisjon
tma4315:2019h:start [2019-10-11]
jarlet [Messages]
tma4315:2019h:start [2019-10-22] (nåværende versjon)
jarlet [Lecture summary]
Linje 37: Linje 37:
 ===== Tentative curriculum ===== ===== Tentative curriculum =====
  
-Fahrmeir et. al. (2013) [[https://​link.springer.com/​book/​10.1007%2F978-3-642-34333-9|(freely available on springer link)]], ch. 2.1-2.4, B.4, 5.1-5.4, 5.8.2, 6, 7.1-7.3, 7.5, 7.7.  This covers ordinary linear and multiple regression (mostly repetition from [[https://​wiki.math.ntnu.no/​tma4267|Linear statistical models]]), binary regression, Poisson and gamma regression, the exponential family and generalised linear models in general, categorical regression (includes contingency tables and log-linear models, multinomial and ordinal regression),​ linear mixed effects models, generalized linear mixed effects models.+Fahrmeir et. al. (2013) [[https://​link.springer.com/​book/​10.1007%2F978-3-642-34333-9|(freely available on springer link)]], ch. 2.1-2.4, B.4, 5.1-5.4, 5.8.2, 6, 7.1-7.3, 7.5, 7.7.  ​We will also use some material from [[http://​www.maths.bris.ac.uk/​~sw15190/​core-statistics.pdf|Wood (2015)]], see below. 
 + 
 +This covers ordinary linear and multiple regression (mostly repetition from [[https://​wiki.math.ntnu.no/​tma4267|Linear statistical models]]), binary regression, Poisson and gamma regression, the exponential family and generalised linear models in general, categorical regression (includes contingency tables and log-linear models, multinomial and ordinal regression),​ linear mixed effects models, generalized linear mixed effects models. ​ 
  
 Also see [[https://​www.ntnu.edu/​studies/​courses/​TMA4315#​tab=omEmnet|the official ntnu course info]]. Also see [[https://​www.ntnu.edu/​studies/​courses/​TMA4315#​tab=omEmnet|the official ntnu course info]].
Linje 64: Linje 66:
 18/9: More on divergence of Fisher scoring algorithm. ​ Model selection.  ​ 18/9: More on divergence of Fisher scoring algorithm. ​ Model selection.  ​
  
-20/9: Theory behind AIC (Wood sec. 4.6).  Poisson regression. ​ Fisher scoring vs. Newton ​Raphosn ​for poisson regression with non-canonical identity link (see R code for further illustrations).+20/9: Theory behind AIC (Wood sec. 4.6).  Poisson regression. ​ Fisher scoring vs. Newton-Raphson ​for poisson-regression with non-canonical identity link (see R code for further illustrations).
  
 25/9: Poisson regression continued. ​ [[https://​folk.ntnu.no/​jarlet/​statmod/​exams/​2015v/​forside-eksamen-en.pdf|Example (Exam in ST2304, 2015, problem 2a-d)]]. ​ Gamma and lognormal regression. ​ Glms in general and IRLS (ch. 5.4, 5.8.2). 25/9: Poisson regression continued. ​ [[https://​folk.ntnu.no/​jarlet/​statmod/​exams/​2015v/​forside-eksamen-en.pdf|Example (Exam in ST2304, 2015, problem 2a-d)]]. ​ Gamma and lognormal regression. ​ Glms in general and IRLS (ch. 5.4, 5.8.2).
Linje 79: Linje 81:
 16/10: Ordinal regression models.  ​ 16/10: Ordinal regression models.  ​
  
-18/​10: ​+18/​10: ​Introduction to mixed models (ch. 7.1 and 7.2)
  
-<color #​c3c3c3>​30/​10:​ No lecture.+23/10: Mixed model continued. ​ ML and REML estimation (7.3).  
 + ​Bayesian interpretation of the restricted likelihood ([[http://​faculty.dbmi.pitt.edu/​day/​Bioinf2132/​previousDocuments/​Bioinf2132-documents-2016/​2016-11-22/​Harville-1974.pdf|Harville 1974]]) and connections between the profile and restricted likelihood. 
 + 
 +25/10: Mixed models continued. ​ BLUPs of the random effects, hypothesis testing and model selection. 
 + 
 +<color #​c3c3c3>​30/​10:​ No lecture.</​color>​ 
 + 
 +<color #​c3c3c3>​1/​11:​ ...Generalized linear mixed models (GLMMs). ​ Laplace approximation. ​ REML for GLMMs....  
 + 
 +8/11: 
 + 
 +10/11:
 </​color>​ </​color>​
 +
 ===== Recommended exercises ===== ===== Recommended exercises =====
  
2019-10-11, Jarle Tufto