Download presentation

Presentation is loading. Please wait.

Published byErin Sharp Modified over 2 years ago

1
1 Michel Tenenhaus & Carlo Lauro A SEM approach for composite indicators building Michel Tenenhaus & Carlo Lauro

2
2 Economic inequality and political instability Data from Russett (1964), in GIFI Economic inequality Agricultural inequality GINI : Inequality of land distributions FARM : % farmers that own half of the land (> 50) RENT : % farmers that rent all their land Industrial development GNPR : Gross national product per capita ($ 1955) LABO : % of labor force employed in agriculture Political instability INST : Instability of executive (45-61) ECKS : Nb of violent internal war incidents (46-61) DEATH : Nb of people killed as a result of civic group violence (50-62) D-STAB : Stable democracy D-UNST : Unstable democracy DICT : Dictatorship

3
3 Economic inequality and political instability (Data from Russett, 1964) 47 countries

4
4 GINI FARM RENT GNPR LABO Agricultural inequality (X 1 ) Industrial development (X 2 ) ECKS DEATH INST Political instability (X 3 ) Economic inequality and political instability

5
5 Building composite indicators 1.Separately for each block (without taking into account the other blocks). 2.For each block, taking into account all the other blocks (multi-block data analysis). 3.For each block, taking into account the causal model (Structural Equation Modelling).

6
6 1. Using SEM for factor analysis Measurement model

7
7 ULS algorithm S = Observed covariance matrix for MV Goodness-of-fit Index (Jöreskog & Sorbum): = PCA when

8
8 First result This solution is not admissible because

9
9 A solution The variance of residual e2 is fixed to a small value

10
10 Result 2 The variance of residual e2 is fixed to a small value:

11
11 Bootstrap Results Regression Weights: Composite indicator

12
12 Principal component analysis with SEM The variance of the residuals are fixed to 0 :

13
13 Result 3 The variance of the residuals are fixed to 0 :

14
14 Bootstrap Results Regression Weights: Conclusion

15
15 Result 4 The variance of the residuals are fixed to 0 :

16
16 Bootstrap Results Regression Weights: Composite indicator

17
17 2. Using SEM for multi-block data analysis

18
18 Result 5 This solution is not admissible because

19
19 Result 6 Var(e2) is fixed to a small value

20
20 Result 7 MacDonald (1996) proposal All Var(e) are fixed to 0:

21
21 Bootstrap Results Regression Weights: Conclusion

22
22 Result 8 MacDonald (1996) Proposals: (1) All Var(e) are fixed to 0: (2) Composite indicator:

23
23 3. Causal model estimation using SEM-ULS

24
24 Result 9 This solution is not admissible because

25
25 Result 10 Var(e2) is fixed to a small value

26
26 Bootstrap Results Regression Weights: Conclusion

27
27 Result 11 Var(e2) is fixed to a small value Composite indicator:

28
28 Bootstrap Results Regression Weights: Conclusion

29
29 Conclusion Agricultural inequality and Industrial development are drivers of political instability Russet hypotheses are validated: Other composite indicators:

30
30

31
31 Conclusion Agricultural inequality above the average Industrial development below the average DICTATORSHIP

Similar presentations

© 2016 SlidePlayer.com Inc.

All rights reserved.

Ads by Google