Presentation is loading. Please wait.

Presentation is loading. Please wait.

WLS for Categorical Data

Similar presentations


Presentation on theme: "WLS for Categorical Data"— Presentation transcript:

1 WLS for Categorical Data

2 SAS – CATMOD Procedure To fit a model using PROC CATMOD
WEIGHT statement – to specify the weight variable Use WLS option at MODEL statement to obtain WLS estimates

3 Data - Response Whether the investigation of the child also involves further investigation of the siblings REVSIB = 0 (No), 1 (Yes)

4 Data – Covariates q1a – relationship to children:
1 – Biological parent 2 – Common-law partner 3 – Foster parent 4 – Adoptive parent 5 – Step-parent 6 – Grandparent 7 – Other

5 Data - Covariates q2a – Gender of the Caregiver:
0 – Female 1 – Male 99 – No response q3a – Age of the Caregiver: 1 – Less than 19 2 – 19 – 21 3 – 22 – 25 4 – 26 – 30 5 – 31 – 40 6 – Over 40 99 – No Response

6 SAS Code Saturated model: proc catmod; weight wtr;
model revsib=q1a|q2a|q3a_age / wls; run; quit;

7 Output The CATMOD Procedure Data Summary Response revsib Response Levels 2 Weight Variable wtr Populations 28 Data Set T2 Total Frequency Frequency Missing Observations 1574

8 Analysis of Variance Source DF Chi-Square Pr > ChiSq Intercept q1a q2a q1a*q2a 4* q3a_age q1a*q3a_age 7* q2a*q3a_age 3* q1a*q2a*q3a_age 2* Residual NOTE: Effects marked with '*' contain one or more redundant or restricted parameters. Q2a – not significant, but has a three-way interaction?

9 Maximum Likelihood Analysis of Variance
Maximum Likelihood Analysis of Variance Source DF Chi-Square Pr > ChiSq Intercept <.0001 q1a 0* . . q2a 0* . . q1a*q2a 0* . . q3a_age 1* . . q1a*q3a_age 7* . . q2a*q3a_age 1* . . q1a*q2a*q3a_age 6* . . Likelihood Ratio NOTE: Effects marked with '*' contain one or more redundant or restricted parameters. Without WEIGHT statement and WLS option – cannot interpret

10 Analysis of Maximum Likelihood Estimates
Standard Chi- Parameter Estimate Error Square Pr > ChiSq Intercept <.0001 q1a # # # # # q2a # q1a*q2a # # # # # q3a_age # # # # Cannot interpret the Estimates

11 Reduced Model Analysis of Variance Source DF Chi-Square Pr > ChiSq Intercept q1a q3a_age <.0001 q1a*q3a_age 7* Residual 0 . . Try model without Q2A – perhaps there’s no interaction between relationship of children and age group of the caregiver

12 Main Effect Analysis of Variance Source DF Chi-Square Pr > ChiSq Intercept <.0001 q1a <.0001 q3a_age <.0001 Residual Try model with Main Effect only

13 Analysis of Weighted Least Squares Estimates Standard Chi- Parameter Estimate Error Square Pr > ChiSq Intercept <.0001 q1a < q3a_age < < Interpret the estimates: negative estimates  those ones are less likely to have investigation done on the siblings

14 Conclusion For cases where the Caregiver is “Adoptive parent”, it is “highly likely” that the siblings will also be investigated For Caregiver between age 22-25, those cases will also likely to have the siblings investigated Intercept  when not much information is observed regarding the caregiver, chances are the siblings will not be reviewed in the case.

15 Questions WLS is more efficient than ML?
Should the records with “no response” be deleted? Is “99” the best code to indicate “no response”? How would the model change if we have less category in each covariates?

16 Thank you 


Download ppt "WLS for Categorical Data"

Similar presentations


Ads by Google