1. ophi hdca ss11 multidimensional stochastic dominance gy
TRANSCRIPT
-
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
1/70
Brief introduction to multidimensional stochastic dominance
Brief introduction to multidimensional stochasticdominance
Gaston Yalonetzky
Oxford Poverty and Human Development Initiative, University of Oxford
OPHI-HDCA Summer School, Delft, 24 August - 3 September
2011.
We are grateful to the World Bank, two anonymous donors andOPHI for financial support
http://find/http://goback/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
2/70
Brief introduction to multidimensional stochastic dominance
Table of contents
Introduction
Stochastic dominance conditions: for univariate to multivariatesettings
Multidimensional stochastic dominance andcomplementarity/substitutability
Relevance for poverty measurement
Tests for univariate and multivariate stochastic dominanceThe test of Barrett and Donald (2003)
Concluding remarks
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
3/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional versus unidimensional
dominance
Stochastic dominance conditions provide an extreme form of robustnessfor ordinal comparisons: if they are fulfilled a comparison is robust to abroad range of parameter values and families of indices.
http://find/http://goback/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
4/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional versus unidimensional
dominance
Stochastic dominance conditions provide an extreme form of robustnessfor ordinal comparisons: if they are fulfilled a comparison is robust to abroad range of parameter values and families of indices.
Multidimensional dominance is relevant for evaluation functions that mapfrom a multivariate space.
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
5/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional versus unidimensional
dominance
Stochastic dominance conditions provide an extreme form of robustnessfor ordinal comparisons: if they are fulfilled a comparison is robust to abroad range of parameter values and families of indices.
Multidimensional dominance is relevant for evaluation functions that mapfrom a multivariate space. E.g. An index of well-being that depends onseveral aspects of wellbeing.
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
6/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional versus unidimensional
dominance
Stochastic dominance conditions provide an extreme form of robustnessfor ordinal comparisons: if they are fulfilled a comparison is robust to abroad range of parameter values and families of indices.
Multidimensional dominance is relevant for evaluation functions that mapfrom a multivariate space. E.g. An index of well-being that depends onseveral aspects of wellbeing.
In unidimensional dominance second and even third orders may beinteresting/relevant.
http://find/http://goback/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
7/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional versus unidimensional
dominance
Stochastic dominance conditions provide an extreme form of robustnessfor ordinal comparisons: if they are fulfilled a comparison is robust to abroad range of parameter values and families of indices.
Multidimensional dominance is relevant for evaluation functions that mapfrom a multivariate space. E.g. An index of well-being that depends onseveral aspects of wellbeing.
In unidimensional dominance second and even third orders may beinteresting/relevant. In multidimensional dominance second and higher
orders are not that easy to interpret.
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
8/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional versus unidimensional
dominance
Stochastic dominance conditions provide an extreme form of robustnessfor ordinal comparisons: if they are fulfilled a comparison is robust to abroad range of parameter values and families of indices.
Multidimensional dominance is relevant for evaluation functions that mapfrom a multivariate space. E.g. An index of well-being that depends onseveral aspects of wellbeing.
In unidimensional dominance second and even third orders may beinteresting/relevant. In multidimensional dominance second and higher
orders are not that easy to interpret.
By contrast, in multidimensional dominance other things matter: thejoint distribution of the variables, as well as how they complement, orsubstitute, each other in their contributions toward the evaluationfunction (e.g. increasing wellbeing).
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
9/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional dominance conditions
In this lecture we are going to review the derivation ofdominance conditions and extend it to multivariate settings.
B f d l d l h d
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
10/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional dominance conditions
In this lecture we are going to review the derivation ofdominance conditions and extend it to multivariate settings.
We will focus on first-order dominance and bivariatedistributions and derive the respective conditions.
B i f i d i l idi i l h i d i
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
11/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional dominance conditions
In this lecture we are going to review the derivation ofdominance conditions and extend it to multivariate settings.
We will focus on first-order dominance and bivariatedistributions and derive the respective conditions.
We will discuss the logic behind some of the conditions andtheir connections to complementarity and substitutabilitybetween variables.
B ief i t d ti t ltidi e si l st h sti d i e
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
12/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional dominance conditions
In this lecture we are going to review the derivation ofdominance conditions and extend it to multivariate settings.
We will focus on first-order dominance and bivariatedistributions and derive the respective conditions.
We will discuss the logic behind some of the conditions andtheir connections to complementarity and substitutabilitybetween variables.
We will discuss why these conditions are also relevant forpoverty assessments.
Brief introduction to multidimensional stochastic dominance
http://find/http://goback/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
13/70
Brief introduction to multidimensional stochastic dominance
Introduction
Introduction: Multidimensional dominance conditions
In this lecture we are going to review the derivation ofdominance conditions and extend it to multivariate settings.
We will focus on first-order dominance and bivariatedistributions and derive the respective conditions.
We will discuss the logic behind some of the conditions andtheir connections to complementarity and substitutabilitybetween variables.
We will discuss why these conditions are also relevant forpoverty assessments.
We will briefly discuss how to test these conditions.
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
14/70
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
Traditional dominance conditions with one variable: an
example
Consider the following wellbeing function: W(x) =xmaxxmin
U(x)dF(x)
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
15/70
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
Traditional dominance conditions with one variable: an
example
Consider the following wellbeing function: W(x) =xmaxxmin
U(x)dF(x)Traditional dominance conditions stem from expressing differences insocial welfare functions as sums of products of derivatives of the
individual welfare functions and functions of the cumulative (or survival)densities. Example:
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
16/70
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
Traditional dominance conditions with one variable: an
example
Consider the following wellbeing function: W(x) =xmaxxmin
U(x)dF(x)Traditional dominance conditions stem from expressing differences insocial welfare functions as sums of products of derivatives of the
individual welfare functions and functions of the cumulative (or survival)densities. Example:
W =
xmaxxmin
U
x(x)F(x)dx
The condition is then: W0 U | Ux
(x) 0 F(x) 0x
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
17/70
Stochastic dominance conditions: for univariate to multivariate settings
Traditional dominance conditions with one variable: an
example
Consider the following wellbeing function: W(x) =xmaxxmin
U(x)dF(x)Traditional dominance conditions stem from expressing differences insocial welfare functions as sums of products of derivatives of the
individual welfare functions and functions of the cumulative (or survival)densities. Example:
W =
xmaxxmin
U
x(x)F(x)dx
The condition is then: W0 U | Ux
(x) 0 F(x) 0x
The dominance condition is usually expressed in terms of distributions.
E.g. ifx :FA(x) FB(x) then we say that distribution A (first-order)
dominates B.
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
18/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
With two variables we integrate by parts the following function:
W =
ymaxymin
xmaxxmin
U(x)f(x, y)dxdy
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
19/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
With two variables we integrate by parts the following function:
W =
ymaxymin
xmaxxmin
U(x)f(x, y)dxdy
Because we have a joint density (f(x, y)) we can integrate using eithercumulative functions:
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
20/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
With two variables we integrate by parts the following function:
W =
ymaxymin
xmaxxmin
U(x)f(x, y)dxdy
Because we have a joint density (f(x, y)) we can integrate using eithercumulative functions:
W =
xmax
xmin
U
x
(x, ymax)Fx(x)dx
ymaxymin
U
y(xmax, y)F
y(y)dy+
ymaxymin
xmaxxmin
2U
xy(x, y)F(x, y)dxdy
Brief introduction to multidimensional stochastic dominance
http://find/http://goback/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
21/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
...or survival functions:
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
22/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
...or survival functions:
W = xmaxxmin
Ux
(x, ymin)Fx(x)dx
ymaxymin
U
y(xmin, y)F
y(y)dy+
ymaxymin
xmaxxmin
2U
xy(x, y)F(x, y)dxdy
Brief introduction to multidimensional stochastic dominance
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
23/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
Notice the appearance of 2U
xy(x, y).
Brief introduction to multidimensional stochastic dominance
S h d d f l
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
24/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
Notice the appearance of 2U
xy(x, y). This cross-partial derivative determines
whether x and y areALEP complementsor ALEP substitutesin theircontributions to U.
Brief introduction to multidimensional stochastic dominance
S h i d i di i f i i l i i i
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
25/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
Notice the appearance of 2U
xy(x, y). This cross-partial derivative determines
whether x and y areALEP complementsor ALEP substitutesin theircontributions to U. Four conditions stem from the two previous equations:
Brief introduction to multidimensional stochastic dominance
St h sti d i e diti s f i i te t lti i te setti s
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
26/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
Notice the appearance of 2U
xy(x, y). This cross-partial derivative determines
whether x and y areALEP complementsor ALEP substitutesin theircontributions to U. Four conditions stem from the two previous equations:
1. A condition for monotonically increasing functions with ALEP substitute
arguments (e.g. 2U
xy(x, y) 0):
x, y : W0U | U
i (x, y) 0
2U
xy(x, y) 0
Fx(x), Fy(y), F(x, y) 0
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
27/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
Notice the appearance of 2U
xy(x, y). This cross-partial derivative determines
whether x and y areALEP complementsor ALEP substitutesin theircontributions to U. Four conditions stem from the two previous equations:
1. A condition for monotonically increasing functions with ALEP substitute
arguments (e.g. 2U
xy(x, y) 0):
x, y : W0U | U
i (x, y) 0
2U
xy(x, y) 0
Fx(x), Fy(y), F(x, y) 0
Notice that x, y : F(x, y) 0 suffices to ascertain W0.
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
28/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
2. A condition for monotonically increasing functions with ALEPcomplement arguments (e.g.
2U
xy
(x, y)0):
x, y : W0U | U
i (x, y)0
2U
xy(x, y)0
Fx(x), Fy(y), F(x, y) 0
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
29/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
2. A condition for monotonically increasing functions with ALEPcomplement arguments (e.g.
2U
xy
(x, y)0):
x, y : W0U | U
i (x, y)0
2U
xy(x, y)0
Fx(x), Fy(y), F(x, y) 0
Notice that x, y : F(x, y)0 suffices to ascertain W0.
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
30/70
Stochastic dominance conditions: for univariate to multivariate settings
First-order conditions for the bivariate case
3. A condition for monotonically increasing functions with ALEPneutral arguments (e.g.
2Uxy(x, y) = 0):
x, y : W0U | U
i (x, y)0
2U
xy(x, y) = 0
Fx(x), Fy(y) 0
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/http://goback/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
31/70
g
First-order conditions for the bivariate case
3. A condition for monotonically increasing functions with ALEPneutral arguments (e.g.
2Uxy(x, y) = 0):
x, y : W0U | U
i (x, y)0
2U
xy(x, y) = 0
Fx(x), Fy(y) 0
Notice that, in this case, it is necessary to test Fi(i) 0 forevery variable (unlike in the previous cases).
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
32/70
First-order conditions for the bivariate case
4. A condition for ALL individual welfare functions that aremonotonically increasing!
x, y : W0U | U
i (x, y)0
Fx(x), Fy(y), F(x, y)0Fx(x), Fy(y), F(x, y) 0
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
33/70
First-order conditions for the bivariate case
4. A condition for ALL individual welfare functions that aremonotonically increasing!
x, y : W0U | U
i (x, y)0
Fx(x), Fy(y), F(x, y)0Fx(x), Fy(y), F(x, y) 0
Notice that, in this case, it is necessary and sufficient to testF(x, y)0 and F(x, y) 0.
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
34/70
First-order conditions for the general multivariatecase
In the general multivariate case, first-order conditions are based on allcumulative and/or survival functions combining all variables.
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
35/70
First-order conditions for the general multivariatecase
In the general multivariate case, first-order conditions are based on allcumulative and/or survival functions combining all variables. They include allmarginal distributions, Fi(i) and all cumulative and survival joints up toF(x, y, ..., w, z) and F(x, y, ...,w, z).
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
36/70
First-order conditions for the general multivariatecase
In the general multivariate case, first-order conditions are based on allcumulative and/or survival functions combining all variables. They include allmarginal distributions, Fi(i) and all cumulative and survival joints up toF(x, y, ..., w, z) and F(x, y, ...,w, z).
In the case of cumulative distributions, the signs of the cross partial derivativesof the functions for which the conditions hold change the following way:Ui>0, Uij0 and so forth.
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
37/70
First-order conditions for the general multivariatecase
In the general multivariate case, first-order conditions are based on allcumulative and/or survival functions combining all variables. They include allmarginal distributions, Fi(i) and all cumulative and survival joints up toF(x, y, ..., w, z) and F(x, y, ...,w, z).
In the case of cumulative distributions, the signs of the cross partial derivativesof the functions for which the conditions hold change the following way:Ui>0, Uij0 and so forth.
In the case of survival functions, all the signs of the cross partial derivatives ofthe functions for which the conditions hold are positive (e.g. Ui>0,Uij>
0,Uijk>
0).
Brief introduction to multidimensional stochastic dominance
Stochastic dominance conditions: for univariate to multivariate settings
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
38/70
First-order conditions for the general multivariatecase
In the general multivariate case, first-order conditions are based on allcumulative and/or survival functions combining all variables. They include allmarginal distributions, Fi(i) and all cumulative and survival joints up toF(x, y, ..., w, z) and F(x, y, ...,w, z).
In the case of cumulative distributions, the signs of the cross partial derivativesof the functions for which the conditions hold change the following way:Ui>0, Uij0 and so forth.
In the case of survival functions, all the signs of the cross partial derivatives ofthe functions for which the conditions hold are positive (e.g. Ui>0,Uij>
0,Uijk>
0).
Hence when conditions on both the cumulative and the survival functions arefulfilled:
x, y, ..., z : W0U | U
i , 3U
ijk, ... 0
Brief introduction to multidimensional stochastic dominance
Multidimensional stochastic dominance and complementarity/substitutability
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
39/70
ALEP relationships and multidimensional dominance
conditions
Why the conditions on cumulative distributions are associated withALEP substitutability and the conditions on survival functions areassociated with ALEP complementarity?
Brief introduction to multidimensional stochastic dominance
Multidimensional stochastic dominance and complementarity/substitutability
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
40/70
ALEP substitution and F(x,y)0
Brief introduction to multidimensional stochastic dominance
Multidimensional stochastic dominance and complementarity/substitutability
http://find/http://goback/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
41/70
ALEP complementarity and F(x,y)0
Brief introduction to multidimensional stochastic dominance
Relevance for poverty measurement
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
42/70
Applying dominance conditions to poverty measurement
Social poverty functions are characterized by:
Brief introduction to multidimensional stochastic dominance
Relevance for poverty measurement
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
43/70
Applying dominance conditions to poverty measurement
Social poverty functions are characterized by:
Being additive and symmetric across individuals:
P(X; z) = 1NNn=1Pn(x; z).
Brief introduction to multidimensional stochastic dominance
Relevance for poverty measurement
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
44/70
Applying dominance conditions to poverty measurement
Social poverty functions are characterized by:
Being additive and symmetric across individuals:
P(X; z) = 1NNn=1Pn(x; z). Higher poverty is worse.
Brief introduction to multidimensional stochastic dominance
Relevance for poverty measurement
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
45/70
Applying dominance conditions to poverty measurement
Social poverty functions are characterized by:
Being additive and symmetric across individuals:
P(X; z) = 1NNn=1Pn(x; z).
Higher poverty is worse.
Monotonicity: Higher xshould not increase poverty:Pn(x;z)
x 0
Brief introduction to multidimensional stochastic dominance
Relevance for poverty measurement
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
46/70
Applying dominance conditions to poverty measurement
Social poverty functions are characterized by:
Being additive and symmetric across individuals:
P(X; z) = 1NNn=1Pn(x; z).
Higher poverty is worse.
Monotonicity: Higher xshould not increase poverty:Pn(x;z)
x 0
So if we define W(X; z) =P(X;Z) then we can apply the abovementioned conditions to poverty comparisons as well!
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
47/70
The test of Barrett and Donald (2003)
This test is a type of Kolmogorov-Smirnov test of
homogeneity of distributions.
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
48/70
The test of Barrett and Donald (2003)
This test is a type of Kolmogorov-Smirnov test of
homogeneity of distributions. The logic is very intuitive: dominance conditions are based on
comparing integrals ofFA(z) with FB(z) for a range ofz,depending on the order of dominance (e.g. for first order, justcompare cumulative or survival distributions).
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
49/70
The test of Barrett and Donald (2003)
There are four possible outcomes (illustrated with first-order):
1. A dominates B:FA
(z
)FB
(z
)z
z|FA
-
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
50/70
The test of Barrett and Donald (2003)
There are four possible outcomes (illustrated with first-order):
1. A dominates B: FA(z) FB(z)z z|FA
-
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
51/70
The test of Barrett and Donald (2003)
There are four possible outcomes (illustrated with first-order):
1. A dominates B: FA(z) FB(z)z z|FA
-
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
52/70
The test of Barrett and Donald (2003)
There are four possible outcomes (illustrated with first-order):
1. A dominates B: FA(z) FB(z)z z|FA
-
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
53/70
The test of Barrett and Donald (2003)
The generic hypothesis subject to test is the following (illustratedwith first-order):
H0 :FA(z) FB(z)z[zmin, zmax]
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
54/70
The test of Barrett and Donald (2003)
The generic hypothesis subject to test is the following (illustratedwith first-order):
H0 :FA(z) FB(z)z[zmin, zmax]
H1 :z[z,min, zmax]|FA(z)>FB(z)
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
55/70
The test of Barrett and Donald (2003)
The generic hypothesis subject to test is the following (illustratedwith first-order):
H0 :FA(z) FB(z)z[zmin, zmax]
H1 :z[z,min, zmax]|FA(z)>FB(z)
Results of these tests require interpretation in order to ascertainany of the four possibilities.
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
56/70
The test of Barrett and Donald (2003)
1. Test:
H0 (FA FB) vs
H1 (FA> FB)
A B or
equal
2. Test:
H0(FB FA) vs
H1(FB
> FA
)A = B
Notreject
A B
Rejec
t
Switch
Not
reject
B A or
crossing
2. Test:
H0(FB FA) vs
H1(FB> FA)
A and B cross
Reject
B A
Not r
eject
Switch
Rejec
t
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
57/70
The test of Barrett and Donald (2003)
The test proceeds as follows (illustrated with first-order):
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
58/70
The test of Barrett and Donald (2003)
The test proceeds as follows (illustrated with first-order):
For a range of z, estimate: FA(z) FB(z)
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
59/70
The test of Barrett and Donald (2003)
The test proceeds as follows (illustrated with first-order):
For a range of z, estimate: FA(z) FB(z)
Estimate the supremum and multiply by a function of the
sample sizes:S= ( NMN+M)1/2supz[FA(z) FB(z)]
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
60/70
The test of Barrett and Donald (2003)
The test proceeds as follows (illustrated with first-order):
For a range of z, estimate: FA(z) FB(z)
Estimate the supremum and multiply by a function of the
sample sizes:S= ( NMN+M)1/2supz[FA(z) FB(z)]S is the statistic we need. We now need to know how likely it
is for this value to appear under the null hypothesis.
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
61/70
The test of Barrett and Donald (2003)
The test proceeds as follows (illustrated with first-order):
For a range of z, estimate: FA(z) FB(z)
Estimate the supremum and multiply by a function of the
sample sizes:S= ( NMN+M)1/2supz[FA(z) FB(z)]S is the statistic we need. We now need to know how likely it
is for this value to appear under the null hypothesis.
There are different procedures to derive the distribution under
the null hypothesis. Barrett and Donald (2003) develop twotypes. We are going to show one of them: a bootstrapmethod.
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
62/70
The bootstrap method in the test of Barrett and Donald
(2003)
1. We pool the two samples of A and B (a total of N+Mobservations) and draw several subsamples with replacement
(R):
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
63/70
The bootstrap method in the test of Barrett and Donald
(2003)
1. We pool the two samples of A and B (a total of N+Mobservations) and draw several subsamples with replacement
(R):2. For each subsample (say 1000) we estimate:
Sr = ( NMN+M)
1/2supz[FA(z; r) FB(z; r)]
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
64/70
The bootstrap method in the test of Barrett and Donald
(2003)
1. We pool the two samples of A and B (a total of N+Mobservations) and draw several subsamples with replacement
(R):2. For each subsample (say 1000) we estimate:
Sr = ( NMN+M)
1/2supz[FA(z; r) FB(z; r)]
3. The p-value is then approximated by:
pA,B
1RRr=11(Sr>S)
Brief introduction to multidimensional stochastic dominanceTests for univariate and multivariate stochastic dominance
The test of Barrett and Donald (2003)
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
65/70
The bootstrap method in the test of Barrett and Donald
(2003)
1. We pool the two samples of A and B (a total of N+Mobservations) and draw several subsamples with replacement
(R):2. For each subsample (say 1000) we estimate:
Sr = ( NMN+M)
1/2supz[FA(z; r) FB(z; r)]
3. The p-value is then approximated by:
pA,B
1RRr=11(Sr>S)
4. If, say, pA,B
-
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
66/70
Concluding remarks
Stochastic dominance conditions are useful to ascertain whether anordinalcomparison is robust to changes in the parameters or members ofa family of evaluation functions.
Brief introduction to multidimensional stochastic dominanceConcluding remarks
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
67/70
Concluding remarks
Stochastic dominance conditions are useful to ascertain whether anordinalcomparison is robust to changes in the parameters or members ofa family of evaluation functions.
When dominance conditions are not fulfilled then the comparison dependson the choice of parameters (e.g. poverty lines, risk/inequality aversion,
etc.).
Brief introduction to multidimensional stochastic dominanceConcluding remarks
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
68/70
Concluding remarks
Stochastic dominance conditions are useful to ascertain whether anordinalcomparison is robust to changes in the parameters or members ofa family of evaluation functions.
When dominance conditions are not fulfilled then the comparison dependson the choice of parameters (e.g. poverty lines, risk/inequality aversion,etc.).
One could restrict the dominance analysis to smaller sets of parameters(or members of families of indices), but this needs to be done carefully,lest significant parts of the domain of interest may be left out.
Brief introduction to multidimensional stochastic dominanceConcluding remarks
C l d k
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
69/70
Concluding remarks
Stochastic dominance conditions are useful to ascertain whether anordinalcomparison is robust to changes in the parameters or members ofa family of evaluation functions.
When dominance conditions are not fulfilled then the comparison dependson the choice of parameters (e.g. poverty lines, risk/inequality aversion,etc.).
One could restrict the dominance analysis to smaller sets of parameters(or members of families of indices), but this needs to be done carefully,lest significant parts of the domain of interest may be left out.
When one is concerned for cardinal comparisons, e.g. between countriesor across time, then dominance conditions are not that useful. Sensitivityanalysis is required.
Brief introduction to multidimensional stochastic dominanceConcluding remarks
C l di k
http://find/ -
7/27/2019 1. OPHI HDCA SS11 Multidimensional Stochastic Dominance GY
70/70
Concluding remarks
Stochastic dominance conditions are useful to ascertain whether anordinalcomparison is robust to changes in the parameters or members ofa family of evaluation functions.
When dominance conditions are not fulfilled then the comparison dependson the choice of parameters (e.g. poverty lines, risk/inequality aversion,etc.).
One could restrict the dominance analysis to smaller sets of parameters(or members of families of indices), but this needs to be done carefully,lest significant parts of the domain of interest may be left out.
When one is concerned for cardinal comparisons, e.g. between countriesor across time, then dominance conditions are not that useful. Sensitivityanalysis is required.
We have assumed the variables are continuous. But these results can beeasily extended to ordinal variables (e.g. see Yalonetzky, 2011). Furtherextensions to combinations of continuous and ordinal variables arepossible.
http://find/