Checking the Independence Assumptions of Generalized Linear Models (GLM)
Learn how to check the independence assumptions of GLM models and ensure model validity. Understand principles such as heterogeneity and repeated measures.
Checking the Independence Assumptions of Generalized Linear Models (GLM)
E N D
Presentation Transcript
Quantitative Methods Checking the models I: independence
Checking the models I: independence Assumptions of GLM
TREATMNT Coef 1 1 BACAFTER = m + bBACBEF + 2 2 + 3 -1 -2 TREATMNT Coef PREDICTED 1 -1.590 BACAFTER = -0.013 + 0.8831BACBEF + 2 -0.726 32.316 Checking the models I: independence Assumptions of GLM BACAFTER = BACBEF+TREATMNT (Model Formula) (Model) (Fitted Value Equation or Best Fit Equation)
TREATMNT Coef 1 1 BACAFTER = m + bBACBEF + 2 2 + 3 -1 -2 Checking the models I: independence Assumptions of GLM BACAFTER = BACBEF+TREATMNT (Model Formula) (Model)
TREATMNT Coef 1 1 BACAFTER = m + bBACBEF + 2 2 + 3 -1 -2 Checking the models I: independence Assumptions of GLM (Model)
TREATMNT Coef 1 1 BACAFTER = m + bBACBEF + 2 2 + 3 -1 -2 Checking the models I: independence Assumptions of GLM (Model) Assumptions of GLM Independence Homogeneity of variance Normality of error Linearity/additivity
TREATMNT Coef 1 1 BACAFTER = m + bBACBEF + 2 2 + 3 -1 -2 Checking the models I: independence Assumptions of GLM (Model) Assumptions of GLM Independence Homogeneity of variance Normality of error Linearity/additivity
Checking the models I: independence Independence in principle
Checking the models I: independence Heterogeneous data
Checking the models I: independence Heterogeneous data
Checking the models I: independence Heterogeneous data
Checking the models I: independence Heterogeneous data
Checking the models I: independence Heterogeneous data
Checking the models I: independence Heterogeneous data
Checking the models I: independence Repeated measures
Checking the models I: independence Repeated measures
Checking the models I: independence Repeated measures
Checking the models I: independence Repeated measures
Checking the models I: independence Repeated measures Single summary approach Multivariate approach Few summaries approach
Checking the models I: independence Repeated measures name C100 ’wtg’ let wtg=LOGWT20-LOGWT3 glm wtg=diet GLM RATE=DIET LET K3=3-31/3 ! 31/3 is the average of LET K8=8-31/3 ! 3, 8 and 20 LET K20=20-31/3 LET K1=K3**2+K8**2+K20**2 LET RATE=(K3*LOGWT3+K8*LOGWT8+K20*LOGWT20)/K1
Checking the models I: independence Repeated measures
Checking the models I: independence Repeated measures GLM LOGWT60 RATE = DIET; MANOVA; NOUNIVARIATE.
Checking the models I: independence Nested data
Checking the models I: independence Nested data
Checking the models I: independence Detecting non-independence In principle: would knowing the error for one or more datapoints help you guess the error for some other datapoint? Experiments: Does the datapoint correspond to the level of randomisation? Observations: Are there groups of datapoints which are very likely to have similar residuals? Be suspicious of - Too many datapoints - Implausible results - Repeated measures
Checking the models I: independence Last words… • Independence is a key assumption, and is the most problematic in practice • Always be alert to possible violations • Know what can be done at the analysis stage • Realise that mistakes at the design stage are often unrecoverable at analysis Checking the models II: the other three assumptions Read Chapter 9