Beruflich Dokumente
Kultur Dokumente
ISLAND
GRADUATE SCHOOL
BOGOR AGRICULTURAL UNIVERSITY
BOGOR
2015
ii
iii
DECLARATION
I declare that this thesis is my original work and that all resources of
materials used for this thesis have been duly acknowledged. This thesis has been
submitted to Graduate School Department of Statistics Bogor Agricultural
University in partial fulfillments of the requirements for M.Sc. degree in
Statistics. I seriously declare that this thesis has not been submitted to any other
institution and anywhere for the award of any academic, degree, diploma, or
certificate.
SUMMARY
MULUGETA AKLILU ZEWDIE.Spatial Econometrics Model of Poverty in Java
Island. Advised by. MUHAMMAD NUR AIDI and BAGUS SARTONO.
Poverty means people's basic needs like food, clothing, and shelter are not
being met. It is generally of two types: Absolute poverty and Relative Poverty:
absolute poverty is synonymous with destitution and occurs when people cannot
obtain adequate resources (measured in terms of calories or nutrition) to support a
minimum level of physical health. Relative poverty occurs when people do not
enjoy a certain minimum level of living standards as determined by
a government (and enjoyed by the bulk of the population).
Keywords: Spatial Effects, Spatial Error Model, Special Lag Model, Spatial
Durbin Model, Robust LM, &AIC
v
RINGKASAN
MULUGETA AKLILU ZEWDIE.Spasial Ekonometrika Model Kemiskinan di
Pulau Jawa. Dibimbing oleh. MUHAMMAD NUR AIDI dan BAGUS
SARTONO.
Kata kunci: Spasial Efek, Spasial Error Model, Spasial Lag Model, Spasial Durbin
Model, Robust LM, dan AIC
vi
Thesis
GRADUATE SCHOOL
BOGOR AGRICULTURAL UNIVERSITY
BOGOR
2015
ix
Accepted by
Adviser committee
Approved by
ACKNOWLEDGMENTS
Thanks and Praise to the Living triune God who guided, provided and
sustained me with wisdom, courage and perseverance throughout this journey.
Next I would like to deeply thank my advisors Dr. Ir. Muhammad Nur
Aidi and Dr. Bagus Sartono for their valuable advice and constructive comments
in while working this thesis, and throughout accomplishment of my entire career
in IPB.
From my heart, I would like to express my gratitude to the government of
Indonesia for financial support during my study in IPB and specially department
of statistics for providing me the opportunity of doing my M.Sc. study.
Finally my deep thanks go to all my family, colleagues and others who
encouraged me in various aspects while conducting this Thesis.
Bogor, Oct.2015
TABLE OF CONTENTS
TABLE OF CONTENTS i
LIST OF ABBREVIATIONS iii
LIST OF TABLES iv
LIST OF FIGURES iv
LIST OF APPENDICES iv
1.INTRODUCTION 1
Background 1
Statement of the Problem 3
Objective of the Study 3
Significance of the Study 3
2.LITERATURE REVIEW 4
Multiple linear regression 5
Weighted Matrix 5
Test Spatial Autocorrelation 6
Spatial Econometric Models 6
Spatial Lag Model 7
Spatial Error Model 7
Spatial Durbin Model 7
Test for Assumptions 10
Model Selection 11
3.RESEARCH METHODS 11
Description of Study Area 11
Data 11
Dependent Variable 12
Independent Variables 12
Steps of Analysis 12
4.RESULTS AND DISCUSSION 14
Descriptive Statistics 14
Global vs Spatial Econometric Model 16
5.CONCLUSIONS AND RECOMMENDATIONS 18
Conclusions 18
Recommendations 18
REFERENCES 19
APPENDICES 20
BIOGRAPHY 22
iii
LIST OF ABBREVIATIONS
DW Durbin Watson
LM Lagrange Multiplier
RI Republic of Indonesia
iv
LIST OF TABLES
LIST OF FIGURES
LIST OF APPENDICES
Appendix 1: R Syntax 20
Appendix 2: Java Island in Map 21
1
1. INTRODUCTION
Background
Poverty is pronounced deprivation in well-being meaning people's
basic needs like food, clothing, and shelter are not being met. Poverty is generally
of two types: (1) Absolute poverty is synonymous with destitution and occurs
when people cannot obtain adequate resources (measured in terms of calories or
nutrition) to support a minimum level of physical health. Absolute
poverty means about the same everywhere, and can be eradicated as demonstrated
by some countries. (2) Relative poverty occurs when people do not enjoy a certain
minimum level of living standards as determined by a government (and enjoyed
by the bulk of the population) that vary from country to country, sometimes
within the same country. Relative poverty occurs everywhere, is said to be
increasing, and may never be eradicated. The measurement and analysis of
poverty and inequality is crucial for understanding peoples’ situations of well-
being and factors determining their poverty situations. A poverty profile describes
the pattern of poverty, but is not principally concerned with explaining the causes
of poverty. Yet, a satisfactory explanation of why some people are poor is
essential if we are to be able to tackle the roots of poverty. Among the key
correlates, of poverty are region-level characteristics, which include vulnerability
to flooding or typhoons, Remoteness, quality of governance, and property rights
and their enforcement. Community-level characteristics, which include the
availability of infrastructure (roads, water, electricity) and services (health,
education), proximity to markets, and social relationships. Household and
individual characteristics, among the most important of which are
a. Demographic, such as household size, age structure, dependency ratio,
gender of head.
b. Economic, such as employment status, hours worked, property owned.
c. Social, such as health and nutritional status, education, shelter.
An economic approach to poverty frequently measures poverty
quantitatively in terms of per capita consumption, income levels or calorific
intakes, such methods used by the World Bank and the UN, Which reflects the
minimum income or consumption necessary to meet basic needs. For low-income
countries, the World Bank has calculated poverty lines between $1 and $2 a day.
Although these minimum requirements vary across countries and over time, $1
and $2 a day measures allow policymakers to compare poverty across countries
using the same reference point. The head count measure of poverty identifies the
percentage of the population living in households with consumption or income per
person below the poverty line. (Coudouel et al.2002).The head count is reported
either as a percentage (the incidence of poverty) or as the number of individuals
who are poor. Another popular measure is the poverty gap, which measures the
mean distance below the poverty line as a proportion of the poverty line.
(Harrison 2007). Poverty studies have, for some time, sought to disaggregate the
poor in order to refine the understanding of causes of poverty and design effective
interventions.
2
Halving the number of people living in less than a dollar a day by 2015 was
the key focus of the Millennium Development Goals agreed to at the Fifty-fifth
session of the General Assembly of the United Nations and subsequently adopted
by leading development institutions. They have inspired unprecedented efforts to
meet the needs of the worlds poorest and have focused attention on the issue of
poverty reduction amongst the most poor. Yet, despite the progress that has been,
made recent estimates suggest that between 300 million and 420 million people in
the world are living in chronic poverty (McKay & Lawson 2003)while currently
this number is increased.
Poverty is one of the fundamental problems that become the center of
attention of the governments of all countries in the world, especially for
developing countries like Indonesia. As National Development Planning Agency
(Bappenas 2010) report peoples living below poverty line in Indonesia is still
quite large. In 2010, Bappenasas figured the numbers of poor people in Indonesia
are around 31.02 million. Additionally, Bappenas noted that as much as half of the
total percent or around 55.83% of the total poor population in Indonesia settled in
Java Island (Bappenas 2010).Java Island is the most populous island in Indonesia.
It consists of 6 provinces namely the Special Capital Region of Jakarta, West
Java, Banten, Central Java, Yogyakarta and East Java. Each province consists of
several districts. One of the efforts made to address the problem of poverty is to
identify the variables that affect poverty on these districts. The poverty of the
district due to the impact of poverty in the surrounding district indicates a spatial
effect. Based on the first law of Geography, “everything is interconnected to each
other, but something close more influence than something far”. (Lee et al.2001).
Indonesia's poverty line is determined by a complex function taking in what
the poor spend on different kinds of food to reach 2.100 calories per day, as well
as costs associated with dozens of non-food goods, including housing, clothing,
education and health care. The poverty line is established as an average, allowing
for the fact that prices vary widely from urban to rural areas, and from more
prosperous Indonesian regions. Based on the government's official poverty line is
233.740 rupiah per capita per month which is close to UN poverty line
measurement from 1-2 dollar a day.
A spatial analysis framework offers advantages over tabular analysis. From
spatial analysis perspective Poverty maps are important tools that provide
information on the spatial distribution of poverty. It used to affect various kinds of
decisions, ranging from poverty alleviation programmers’ to emergency response
and food aid. The Visualization of the estimates in map form is an efficient
medium for planning responses to poverty. Spatial statistics can quantify and
clarify patterns seen in maps. A spatial framework allows for incorporating
spatially continuous environmental variables in the analysis. Explicit spatial
analyses take into account the local nature of relationships between poverty and
its determinants. However, the use of poverty maps alone does not furnish an
estimate of the causal linkage between poverty and the variables influencing it;
such maps furnish only “visual” advice. For this reason, researchers usually look
for the possible existence of empirical relationships between poverty and socio-
economic indicators that was applied in South Africa and Ecuador (Hentschel et
al.2000).
3
Objectives
Main objective
The main objective of this research is to identify the variables that affect
poverty by applying Spatial Econometric Model in Java Island with105 districts.
Specific Objective
a. To identify the variable that significantly determine poverty
b. To make policy recommendation to prevent and alleviate poverty
c. To compare the best model among classical linear regression and Spatial
Econometric Models
2. LITERATURE REVIEW
Nelson Mandela came out of retirement in February 2003 to speak on behalf
of to Make Poverty History campaign in London, an effort to renew the global
commitment to eliminating poverty worldwide. “Like slavery and apartheid,
poverty is not natural”, Mandela intoned. “It is man-made, and it can be overcome
and eradicated by the action of human beings.” In imagining a world without
poverty, hope that Mandela’s strong voice will spur surer action to eliminate the
deprivations suffered by the world’s poor.
Statistical offices spend much time and effort setting and updating poverty
lines. However, the place of poverty lines needs to be put in context. A recent
study of 17 Latin American countries, for example, shows that many other
elements of poverty measurement are more important than the choice of poverty
lines. These include adjustments for adult equivalent family size and the treatment
of missing data in surveys is important rather than choosing poverty line. (Szekely
et al. 2000).
Over the past many years, the causes and consequences of poverty, and
changes in poverty over time, have been the subjects of much academic research
and social policy debate. In large measure, two schools of thought have dominated
this research and debate. One attributes the causes of poverty primarily to
individualistic or family compositional forces. Sometimes referred to as “people
poverty”, another school of thought focuses on contextual or structural forces,
sometimes referred to as “place poverty.” These include issues such as urban
economic dislocations, faltering regional economies, high unemployment, poor
and often disorganized local employment opportunity structures all forces over
which the individual has little or no control. (Pebley & Sastry 2003).
A key element affecting poverty is regionalism; Said Levernierand
concluded that economic development targeting predominantly African-American
community’s counties would be most effective in alleviating poverty.Triest 1997
concluded that increased employment of the low-income population and increased
educational opportunity would narrow the interregional gap in poverty. Goetz
2007 suggested that government can increase investment in social capital to
reduce the poverty rate by easing transaction costs paid by local
associations.Findeis found that welfare assistance to help the poor workers had
effects on poverty in metro areas. Mauro found that the poor countries tend to
have corrupted bureaucracies and politic instability. (McKay & Perge 2011)
5
Where:
y i : poverty rate of the i-th district (i = 1,2,…,n)
: constant term or y intercept
0
j : Regression parameters(j=1,2…k)
x ij : predictor variables
i : Random error term iid with mean zero and constant variance.
Weighted Matrix
The basis for most models is an indicator of whether one region is a spatial
neighbor of another; or equivalently, which regions are neighbors of a given
region. This is a square symmetric weighted matrix (W) nxn (row standardized)
matrix that define who are neighbours with who. To construct this weighted
matrix:
1) Contiguity based weighted matrix: Queen
2) Inverse Distance weighted matrix
3) K –nearest neighbor weighted matrix
For this Research the researcher used contiguity weighted matrix that is
based on queen two regions are neighbors in this sense if they share any part of a
common border, no matter how short is it. Queen Contiguity weighted matrix is as
follows:
6
w
j1
ij
(n 1)(W )
2 2
ij
ij
ε ~ N (0, 2 I)
Since in the likelihood ratio common factor test according to Elhorst’s flow
chart if the null hypothesis accepted lamda will be equal to rho so the model can
be write in above form.
Under SDM by using likelihood ratio common factor we can come out to
SEM model & SAR model.H0: if it is not significant our model goes to
SEM.if it is significat it will be SAR or OLS depending on spatial lag coeeficients
here is the concept:
The entire above all models uses Maximum likelihood estimation for the
parameters beta and teta.
Xj Xi Xj Xi Xj Xi Xj Xi
Yj Yi Yj Yi Yj Yi Yj Yi
Ɛj Ɛi Ɛj Ɛi Ɛj Ɛi Ɛj Ɛi
No influence Dependent variable Residuals Independent variable
from neighbors Influenced by influenced by influenced by
neighbors neighbors neighbors
tr (W W W )
t t
The next Lagrange Multiplier Test for Spatial Lag (LM-Lag) hypothesis is :
H0: = 0 (there is no spatial lag effect)
H1: ≠ 0 ( there is spatial lag effect )
The Lagrange Multiplier test statistics for this is:
LM lag [e tWy / e t e / n)]2 /[ D tr (W 2 W tW ]
Where D [(WX ) t ( I X ( X t X ) 1 X t )(WX ) /(e t e / n)]
Reject Ho LMlag> ( ,1)
2
Another test used is Likelihood ratio common factor test with the following
hypothesis: H0: vs H1: With Test statistics of :
V (εj ) = σ2 for all j. That is, the variance of the error term is constant.
(Homoskedasticity). If the error terms do not have constant variance, they are said
to be heteroskedastic. [Tidbit from Wikipedia: The term means “differing
variance” and comes from the Greek “hetero” ('different') and “skedasis”
('dispersion').]Breusch pagan test is one of the test statistics to test it and
Hypothesis are as follows: H0: Homoscedasticity H1: hetroscedasticity
With Test statistics where SSR(explained sum squared /sum squared
dueto regresion from new model) Ri=Xi hence refer the detail
of this equation written by Arbia (2006) . Reject H0 where n is number
of independent variable.
One assumption of our model is that the error terms are independent. In this case
if the order of autocorrelation more than one Moran Index test would be better.
The Durbin-Watson test statistic is typically used to test (there is no
autocorrelation) ρ (residuals areautocorrelated),where the test statistics
is as follows:
e ei 1
n 2
i 1 i
d cal Reject Ho dcal ≤ dL,α/2 or dL,α/2 ≤ (4 – dcal) ≤ dL,α/2
n 2
e
i 1 i
11
Model Selection
Model selection can be helpful to identify a single best model or to make
inferences from a set of multiple competing hypotheses Up to now, however, only
a few model selection procedures have been tested for spatially auto correlated
and spatial lag data. Therefore the researcher developed model selection
procedures and selected the best models among OLS, SDM, SAR and SEM by
model selection criteria of akaike information criteria (AIC) as follows:
3. RESEARCH METHODS
Data
The data is secondary data which is collected by BPS Indonesia, in 2010.
All the response and explanatory variables are continuous and all variable are
changed to percentage for analysis purpose.
Dependent Variable
The responses variable in this study is percentage of poverty rate at
particular city or region of 105 districts.
12
Independent Variables
The explanatory variables that are included in this study by assumed to be
correlates to poverty are:
X1: Percentage Unemployment rate
X2: Percentage Malnutrition rate
X3: Percentage Child mortality rate
X4: Percentage Morbidity (occurrence of disease)
X5: Percentages of household more than high school
X6: Percentage of access to clean water
X7: Percentage of non-sanitation
X8: Percentage Literate rate
X9: Percentage Employment rate
X10: Percentage of un worked hour per week
X11: Percentage health complain of the household
X12: Length of sickness
Methods of Analysis
1. Data exploration with graph and descriptive statistics: this part explain
descriptive part of the analysis by using bar chart of the given province data
without considering the districts so that can know the general outline of poverty in
java island.
2. Analysis Multiple linear Regression model using OLS estimation: In this part
all we want to compare the global model by using ols estimation after that we
want to extract the error to check the assumption.
3. Create row standardized weighted matrix Using contiguity Queen Criteria: the
row standardized matrix helpful before go to the spatial part must be created here
by looking Queen Criteria refer more detail for this matrix in the Research
reviews.
4. Test for the existence of spatial autocorrelation using Moran Index test :
After we get the weighted matrix we want to know about the spatial correlation
since spatial correlation can affect the result of multiple linear regressions so that
Moran I test will be held on here.
5. Test for spatial lag and spatial error effect by using LM and Robust LM test
If the spatial effect occurs we need to identify the source of its effect by using
Lagrange multiplier Test.
6. Analysis Spatial Lag Model, Spatial Error Model and SDM model
The analysis our Spatial Econometrics model so far we already identified the
effects of spatial models in the existence of autocorrelation
7. Under spatial Durbin model test LRcom factor test and come up to the reduced
model: likelihood ratio common factor test is very important to identify our best
model.
8. Select the best model by using model selection criteria and test assumptions of
residual: Finally we used model selection criteria that are mention in the research
review.
13
Data Exploration
Analysis OLS
Diagnosis Analysis SDM
(LMerr&LMlag)
All significant
Significant
AIC
16.83%
15.26% 15.56%
11.27%
7.16%
3.48%
DKI Jakarta Banten West Java East Java Central Java DIY
Yogyakarta
Figure 3. Percentage of poverty rate in Java Island.
From the above figure we can see that DKI Jakarta, Banten, West Java,
East Java, Central Java and DIY Yogyakarta indicated by depending up on poverty
rate respectively. So that among the Java Island provinces in 2010 house hold
survey there was high percentage of poverty rate in DIY Yogyakarta, Central Java
and East Java respectively while in DKI Jakarta is relatively small percentage of
poverty rate. In the next step of finding the researcher look the factors that affect
poverty rate in Java Island. Why poverty rate is less in DKI Jakarta some
researcher found that more rural places are worse in poverty than urban why we
shall get it on the outcome. The next step after it is creating weighted matrix with
105 by 105 matrixes after that test the existence of spatial autocorrelation and
spatial dependency. If there is no spatial autocorrelation keep our classical linear
regression model and give conclusion and recommendation based on it.
15
Table 1 Spatial autocorrelation test and spatial dependence test for poverty of ols
Moran Index Test for autocorrelation
Moran I statistic standard deviate = 8.464, p-value < 2.2e-16
Moran I statistic Expectation Variance
0.5596 -0.0096 0.0045
Lagrange multiplier diagnostics for spatial dependence
LMerr = 19.016, df = 1, p-value = 1.296e-05
LMlag = 27.620, df = 1, p-value = 1.477e-07
Robust Lagrange multiplier diagnostics for spatial dependence
RLMerr = 0.0015, df = 1,p-value = 0.9691
RLMlag = 8.6051, df = 1, p-value = 0.0034
From the above table 1 Moran I test statistics (0.56) indicated that there is
a positive autocorrelation in this poverty data. And the researcher test the
significant of autocorrelation by looking p value (2.2e-16) that is very small and
less than 0.05 so reject the null hypothesis as stated in the research methods and
we conclude that there is a positive spatial autocorrelation in the given poverty
data meaning high values of a poverty rate at one locality are associated with high
values at neighboring localities or low values of a poverty rate at one locality are
associated with low values at neighboring localities since the spatial
autocorrelation is positive. In another way Moran’s I (0.56) can be interpreted as
the correlation between variable, poverty rate, and the spatial lag (Wy) of poverty
rate formed by averaging all the values of poverty rate for the neighboring
polygons. Now after the existence of spatial autocorrelation the researcher needs
to test spatial dependence, if spatial autocorrelation exist spatial dependence will
also exist. First, check the significance of the Lagrange Multiplier (LM) test,
which tests for the presence of spatial dependence. If only one is significant, (lag
or error), proceed to do that test. If both are significant, check the Robust LM
tests, which tests which one could be at work. If only one is significant in Robust
test, (lag or error), then do that test. If they are both significant, choose the test
with the biggest value. From Lagrange multiplier as we seen all spatial lag and
spatial error dependence occur so our model should not be OLS so far we also
know that as spatial autocorrelation occur spatial dependence and spatial
heterogeneity also occur how ever in the lm test both of the error and the lag
model will be appropriate but Anselin stated that we should go further robust lm
test so as we can see the error model is no more significant than the lag model in
Robust LM test. So from here we can say our best model well be SAR model
while latter will see on LR test and AIC as comparison.
16
From the above table 2 the researcher can observe that the more
appropriate model for our poverty data is the spatial lag model which has
minimum AIC (43.8) even the likelihood ratio common factor test pointed that
spatial Durbin model is differ from spatial error model. If spatial Durbin model is
differ from spatial error model or it cannot be reduced to spatial error model so
that our model pointed to OLS or Spatial Lag Model; in above table as discussed
17
before the OLS result is affected by the presence of spatial dependence and even
seen unexpected sign since the spatial effect are significant so that the best model
is spatial lag model. So from it as we can see that literate rate and house hold who
has higher education is a negative impact on poverty while employer who has
more un worked hours per week has a positive impact on poverty. In our lag
model the spatial lag effect is significant (0.51702) which mean that on average
100 percent increased in poverty rate in a location resulted in 51.7 percentage
point increase in poverty rate in neighbors location and the highest significant of
error lag also indicated that a random shocked in spatially omitted variable that
affects percentage of poverty rate in a particular location triggers a change in
percentage of poverty rate. The next thing is to check our best model to fulfills the
requirement of assumptions remember the dependent variable was change to log
since the residuals have a skewed distribution. The purpose of a transformation is
to obtain residuals that are approximately symmetrically distributed (about zero,
of course),all the non bracket page with respect to each model shows the value of
the variable is insignificant and under bracket of above table shows p value.
Table 3. Assumptions Test SAR/lag model
No. Tests Test Process Stat vs DF P-value vs CI
1 Normality KS-Test 0.095238 0.7277
2 Homogeneity Breusch- 11.043(11) 0.4397
Pagan test
3 Independency DW-Test VS 2.24 (dl=1.416,
Moran I test du=1.948)
From the above table 3, residuals of the lag model as we seen it, it is
enough to say that our model has no problem on normality assumption, From KS
test we can also conclude that our model is normal distributed since
(KS=0.095238 with p-value=0.7277) indicated that we accept the null hypothesis
so that there is no normality problem in our model. For more the researcher also
tested the constant variance assumption here the result above from BP test
indicated that there is no more heterogeneity problem since the p value is greater
than 0.05 we accept the null hypothesis that mean the variance is homogen.
Remember as stated before in the research methods our null hypothesis is
homoscedasticity against hetroscedastcity.The OLS result of Durbin Watson is
1.56 which indicates that residual are auto correlated so that the OLS model will
not accurate since this assumption violated while the SAR model as we seen
above table DW=2.24 is greater than du that means there is no problem of
autocorrelation or don’t reject the null so that our model is good enough, it is also
checked by Moran since Dw test only test the first autocorrelation effect.
Remember multicollinearity also checked using VIF.
18
Recommendation
As individual level the researcher recommend to the household of all family
member must be increase there working time if they are whatever government
employer or private employer so that it can help to generate income and alleviate
poverty.
As a government level the researcher recommended that the policy must
focused on developing human capacity by increasing literacy rate and education
should be free and supported by government until strata one so that educated
people can be alleviate poverty in many direction.
Due to limitation of resource the researcher does not cover all the expected
factors of poverty.
19
REFERENCES
Anselin L.1988. Spatial Econometrics: Methods and Model, Kluwer, Dordrecht.
Anselin L.2001.Spatial Econometrics in a companion to Theoretical Econometrics
( Baltagi B.H. ed). Blackwell.Oxford.
Anselin L.2010.Lagrange multiplier diagnostics for spatial dependence and
heterogeneity, Geographical Analysis. Wiley online library.
Arbia G. 2006.Spatial Econometrics Statistical Foundations and Applications to
Regional Convergence, Italy.
[BAPPENAS] National Development Planning Agency. 2010. Report on the
Achievement of the Millennium Development Goals Indonesia.
Coudouel A, Jesko H, Quentin W. 2002. Poverty Measurement and Analysis, in
the PRSP Sourcebook, World Bank, Washington D.C.
Harrison A. 2007.Globalization and Poverty, NBER Books, National Bureau of
Economic Research.
Hentschel J, Lanjouw P, Poggi J. 2000. Combining census and survey data to trace
the spatial dimensions of poverty: a case study of Ecuador. World Bank
Econ.
Higazi SF, Abdel-Hady DH, Al-Qulfi SA. 2013. Application of Spatial Regression
Models to Income Poverty Ratios in Middle Delta Contiguous Counties in
Egypt.Tanta University, Tanta, Egypt
Lee J, David W. 2001. Statistical analysis with arc view GIS, John Wiley, New
York.
Lesage JP, Pace K. 2009. Introduction to Spatial Econometrics, Boca Raton: CRC
Press.
Lesage JP. 1999. The Theory and Practice of Spatial Econometrics, Department of
Economics University of Toledo.
Mckay A, Lawson D. 2003.Assessing the Extent and Nature of Chronic Poverty in
Low Income Countries: Issues and Evidence, University of Nottingham,
UK.
McKay A, Perge E. 2011. How strong is the evidence for the existence of
poverty traps? A multi country assessment, Working Paper series.
Mur J, Angulo A. 2006. The Spatial Durbin Model and the Common Factor Tests,
Spatial Economic Analysis.
Paraguas FJ, Kamil A. 2005. Spatial Econometrics Modeling of Poverty paper
presented on the 8th WSEAS International Conference on applied
mathematics, Tenerife, Spain.
Pebley AR, Sastry N. 2003. Neighborhoods, Poverty and Children’s Well-being,
University of California, Los Angeles.
Rawling JO, Pantula SG, Dickey DA. 1998. Applied Regression Analysis A
Research Tool Second Edition. Raleigh, North Carolina USA.
Szekely M. Lustig N, Cumpa M, Mejia JA. 2000.Do We Know How Much
Poverty There Is? Inter-American Development Bank, Felipe Herrera
Library.
20
APPENDIX
A1. R-Syntax
poverty<-read.table("D:/poverty14.csv",sep=",",header=TRUE)
weight<-read.table("D:/weight12.csv",sep=",",header=FALSE)
attach(weight)
attach(poverty)
w<-as.matrix(weight)
library(spdep)
library(lmtest)
mat2listw(w)
moran.test(poverty$y,mat2listw(w))
ols<-lm(y~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12)
summary(ols)
bptest(y~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12, data=poverty)
lm<-lm.LMtests(ols,mat2listw(w),test=c("LMerr","LMlag"))
lm
sar<lagsarlm(y~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12,data=pov
erty,mat2listw(w)
summary(sar)
bptest.sarlm(sar)
l<-as.matrix(residuals(sar))
sem<errorsarlm(y~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12,data=p
overty,mat2listw(w))
summary(sem)
bptest.sarlm(sem)
e<-as.matrix(residuals(sem))
mod.sdm <- lagsarlm(y~x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12,
data = poverty, mat2listw(w), zero.policy=T, type="mixed",
tol.solve=1e-12)
summary(mod.sdm)
bptest.sarlm(mod.sdm)
d<-as.matrix(residuals(mod.sdm))
durbin.test<-LR.sarlm(mod.sdm,sem)
print(durbin.test)
21
Capital City of RI
BIOGRAPHY
The author of this Thesis entitled “Spatial Econometrics Model of
Poverty in Java Island” was born in 19 May 1989 in Ethiopia. Then, He joined
Department of Statistics at Hawassa University in 2008/9. After finishing of his
first degree in 2011 he employed at Mekelle University as Graduate Assistance
Lecturer after 2 year he joined MSc program in Statistics at Bogor Agricultural
University in RI.