Diklat : T. LN.

01
No. : 01
Nama : Windhiarso Panco Adi Putranto
Nama Pemimbing : Takashi Yahagi Professor
Judul Buku : Raining Of Wavelet Networks Using the Unscented Kalman Filter
for Predicting Nonlinear Time Series
Abstrak :
Time series prediction is takes an existing series of data Xt-n, .. Xt-2, Xt-1, Xt and predict
the Xt+1, Xt+2, …….., data values. The goal is to observe or model the existing data series to
enable future unknown data values to be predicated accurately. Examples of data series
include financial data series, physically observed data series and mathematical generated
data series.
Many techniques have been implemented to perform time series prediction. This thesis
will focus on wavelet networks. The idea of combining both wavelets and neural
networks has resulted in the formulation of wavelet networks – a feed-forward neural
network with one hidden layer of nodes, whose basis function are drawn from a family of
orthonormal wavelets. The use of wavelet network have been applied to speech
segmentation, speaker recognition, face tracking, forecasting, and prediction of chaotic
signals. The problem is how to make an algorithm for determining the parameter values
of the wavelet networks and training the network to adjust the parameters of the network
to minimize some function (usually the square error between the output of the network
and the desired aoutput). The usual method to do this in the domain of neural networks is
the backpropagation. A large fraction of recent work in artificial neural network uses
multilayer perceptrons training with the backpropagation algorithm converges slowly for
large or complex problems such as speech recognition, where thousands of iterations may
be needed for convergence even with small data sets.
The Extended Kalman Filter (EKF) is a well known tool for recursive parameter
estimation, machine learning applications, nonlinear system identifications and for the
purpose of training networks. In particular, the EKF has been applied to the estimation
and weights training of feedforward and recurrent neural network models. Wan and van
der Merwe point out the underlying assumptions and dramback in the EKF, which can
introduce large errors and may lead to suboptimal performance. They presented an
alternative filter with performance superior than the EKF called the unscented Kalman
filter (UKF). This new method improve the drawbacks in the EKF, and perform a better
result on nonlinear estimation problem, including nonlinear system identification and
training of neural networks.
In this thesis, we show that training wavelet networks can be solved using the
unscented Kalman filter. Simulation result are demonstrated to validate the ability and
efficiency of the propose network.
Diklat : T.LN.2004.02
No Buku : 02
Nama : Toto Silitonga
Nama Pembimbing : Prof. Dr. Hidetomo Ichihashi
Judul Buku : A Fuzzy Clustering With Karnel Trick
Absrak :
Fuzzy e- Means clustering algorithm is the popular clustering technique by the distance-
based objective function method. By the addition of a regularizer and the kernel trick to a
fuzzy counterpart of Gaussian mixture density models (GMM). This paper proposes a
clustering al-gorithm in an extended high dimensional feature space.

Unlike the global nonlinear approaches, GMM or its fuzzy counterpart is to model
nonlinear structure with a collection, or mixture, of local linear sub-models of PCA.
When the number of feature vectors and clusters are n and C respectively, this kernel
approach can find up to C x n nonzero eigenvalues. A way to control the number of
parameters in the mixture of probabilistic principal component analysis (PPCA) is
adopted to reduce the numberof parameters.
The algorithm provides a partitioning with flexible shape of clusters in original input data
space.

Diklat : T.LN. 2004.03

No Buku : 03
Nama : Zuraini
Nama Pembimbing : Prof. Harunori Shishido
Judul Buku : Indonesia’s Rice Problem In 1998 Statistical data And Game
Theory Analysis
Absrak :
This work deals with the rice problem in Indonesia. It was motivated by the inability of
domesrtic production to meet domestic demand. The rice situation including the
distribution system and the 1998 rice rising are tried to describe in this paper.

This thesis uses two tools of analysis i.e. statistical and game theory analysis. Based on
statistical data the paper finds that the increasing number of rice consumer is not all
related to the increasing of population. Increasing of rice consumer’s factor also exists.
Game theory analysis, using consuming rice game, finds an equilibrium with good
outcome, although it is difficult to realize the equilibrium.

Keywords : rice production, rice consumption, increasing of rice consumer, rice

distribution, rice price rising in 1998, consuming rice game.
Diklat : T.LN.2004.04
No Buku : 04
Nama : Agus Setiawan
Nama Pembimbing : Prof. Tomoyuki T ARUMI
Prof. Koji KURIHARA
Judul Buku : Small Area Statistics As A Method And Its Application In
WinBugs Software
Absrak :
Small area estimation is the process of using statistical models to link survey outcome
variables to a set of predictor variables known for small domains, in order to predict
domain-level estimates. The need for detailed statistics on small area is constantly
increasing. Small area estimation is becoming important in survey sampling due to
agrowing demand for reliable small area statistics from both public and private
sectors.Bayesian hierarchical models provide a convenient framework for disease
mapping and geographical correlation studies. Computation may be carried out using the
freely-availeble WinBUGS software. Two Approaches prediction to estimate total patient
in small area i will be presented. For the purpose of this paper, the small area estimation
in this context use data of Indonesia’s population based on the 2000 census for the
population of Jakarta and data of patient diarrhea from District Health Service of Jakarta.
We interest to predict total patient diarrhea as variable of interest and data population as
auxiliary data from unsample for each small area.
Small Area Statsitics (SAS) is commonly conceived of as statistics about such units as
individuals, households, and farms, associated with the small (administrative) areas into
which a larger (administrative) area has been divided. As an example, the small areas
may be counties and the large area a state.

Key words : Auxiliary data, Populations density, Sample survey, Small area estimation,
WinBugs.

Diklat : T.LN2005.05
No Buku : 05
Nama : Yeshri Rahayu
Nama Pembimbing : Dr. Leon Van der Dool,
Prof. Dr. Ir. Achmad Djunaedi, MUP
Judul Buku : The Role Of Upik As One Of Public Participation in Local
Development Process (A Case Study in Yogyakarta Municipality. Indonesia)
Absrak :
He development in Information Communication and Technology (ICT) gives an
opportunity to access, manage and make efficient use of fast and accurate information.
Utilization of ICT changes the bureaucratic culture as a result of new approach in public
management. Governments are there shifting from paperbased administration to a much
faster system of electronice in decision making process. The implementation process of e-
government in Indonesia is supported by President Instruction number 3 in 2003. The law
requires local government to improve and develop e-government in providing better
services to community in order to actualize good governance.

Government of Yogyakarta Municipality (GOYM) is one of municipals in Indonesia

which is intended for the implementation of e-government. Since September 2003, in
order to increase public participation by utilizing e-government, GOYM established
forum to accommodate public aspiration, which is UPIK (Complaint and Information
Service Centre). The focus of this study is to analyze the effectiveness of the program to
accommodate and respond towards the community problems and impact of the program
on local development process. The assessment process conducted by this research will
focused on the UPIK organization, related bureaus and involved community.

It is a case study which used inductive research methodology. Data analysis was based on
combination of qualitative and quantitative data collected during fieldwork. The
assessment process was done by conducting deep interview, questionnaires, and desk
study research of related theory and law that regulate the organizations.

The existence of UPIK in GOYM takes an important role in development process. Due to
limitations on infrastructure, human resources, and accessibility to the program
influenced the effectiveness of the program and level of community participation through
this media. The effect of on line services to development process helped bureaus to
provide better services to community. The real impacts were significantly on bureaus
which directly relates to physical services. The improvement on infrastructure and
resources is needed in order to provide better services.

Keywords : Public participation, ICT, e-government, public services, public management

service

Diklat : T. LN 2007. 06
No Buku : 06
Nama : Edi Setiawan
Nama Pembimbing : Youdi Schipper Phd
Judul Buku : Determinants of Children Schooling Participation in Indonesia
Absrak :
This study intends to find the determinant of children school participation. To do so I try
to combine three main factors. Children, household-parent and region-policy factors.
With children data I examine the age. Sex, house activity and birth order, as with
household-parent data I examine the parent education, income, marital status and
occupational status of household and finally with region-policy data, I examine location,
region, availability of school and subsidy. The main findings are doing house working
activity in the household gives a significant impact on schooling. Income of the family
has an important role in determining the children schooling. Government policy such as
subsidy is very important to increase the probability of schooling.

Aeknowledgements

My supervisor Youdi Schipper, Phd at the Faculty of Economics and Business

Administration (FEWEB) of Vrije University Amsterdam and Dr. Aico Van Vuuren, my
econometric lecture have given me a lot of advice and help. I am really grateful for their
contributions. Also I want to thank my fellow students who gave me a lot of support and
advice during my study. Last but not least I want to thanks to my family who contributed
in many ways to make it possible to finish this study.

Diklat : T.LN.2007.07
No Buku : 07
Nama : Karmaji
Nama Pembimbing :
Judul Buku : Regional Economic Growth and Convergence In Indonesia
Absrak :
This thesis describes the long-term history of national economic growth and its
performance, regional economic growth and regional economic convergence in
Indonesia. It also provides a space and time exploratory description regarding the
distribution of per capita regional income and a space and time confirmatory spatial
analysis of regional economic convergence.
The first issue was to test the hypothesis that regencies with high (respectively low)
income per capita are more spatially clustered than would be due to a scatter arrangement
or by pure random chance. It was purely descriptive when two statistical test for spatial
autocorrelation to identify regional clusters and outliers we employed. The recently
developed method of exploratory spatial data analysis provided new insights on the
geographical dynamics of regional income growth patterns over the 1983–2003 period.
Regional data from 1983-2003 also provided evidence of global andlocal spatial
autocorrelations, indicating the locations of the centers of economic growth that are
spatially aggregated in certain geographic areas rather than simply being scattered
randomly across country and time.
The second issue was related to regional esonomic convergence which is reconsidered
from a spatial econometric perspective. It was found that regional data was subject to
substantial spatial effects during the study period and an econometric analysis of the
conventional convergence model revealed strong evidence of misspecification or
unreliability due to biased estimates of the parameter’s variance, and also biased or
inconsistent estimates for the coefficients due to simulataneity between error terms and
spatially legged dependent variables. Consequently, in the case of Indonesia, it is
necessary to employ spatial econometric modles when analyzing regional economic
convergence that appropriately consider spatial; interdependence among regencies.
Diklat : T, LN.2007.08
No Buku : 08
Nama : Puji Agus Kurniawan
Nama Pembimbing : Yasuhide Okuyama ,
Takahiro Akita
Judul Buku : Regional Income Inequality and Growth Convergence : A Case
Of Indonesia
Absrak :
In this study, We examine regional income inequality and GDP per capita growth
convergence during the period from 1983-2004. We examine regional income enequality
using standard deviation of natural logarithm of GDP per capita, the Williamson’s
weighted coefficient of variation (CVx), and twoTheil indiees while we examine the GDP
per capita growth convergence using cross-section and panel data regression. We find
that regional income inequality as measured by all indices both including and excluding
the mining sector show a declining trend the period, however, all indices including the
mining sector were decreasing more rapidly than all indices excluding the mining sector.
Futhermore, when the mining sector is not taken into account, we observe that the tertiary
sector is considered to be an enequality-increasing sector while the primary sector is
considered as an inequality-decreasing sector. On the other hand, a regional
decomposition shows that 90 percent of the inequality is contributed by enequality
within-region. Meanwhile, the convergence analysis shows that during the period from
1983-2004 the GDP per capita growth converge, meaning that poor provinces grew faster
than richer ones and we also observe that physical capital accumulation, human capital
accumulation, and trade openness has significant effect on GDP per capita growth.

Keyword : Regional Income Inequality, Convergence, Indonesia

Diklat : T, LN.2008.10
No Buku : 10
Nama : Lia Ermayati
Nama Pembimbing :
Judul Buku : Alternative Design of Farmers Terms of Trade Index
Absrak :
The Frmers’ Terms of Trade (FTT)index plays important roles for guiding policy
formulations in the agricultural sector. Currently, its main aim is to measure changes in
farmers’ welfare by covering, among other prices, their purchasing process of
consumption goods/services. The paper examines whether this is the right objective in
consideration of extensive implications of welfare as against limited converage of
consumtions goods/service. Also, its design including sampling, frequency and covered
crops need redesigning in order to improve reliability under the budget constraint. As a
result, recommendations are presented : (a) the FTT should aim to measure changes in
profitability of farm production instead of farmers’ welfare; (b) the FTT should be nade
for each selected major crop produced in each province; (c) the FTT should be made a
few times a year, perhaps around harvesting seasons of each crop, instead of every mont;
and (d) the sample size should be determined in such ways that it will minimize sampling
errors, though at present the level of confidence varies from crop to crop and from
province to province due to widely verying sample sizes, which in many cases are not
large enough.

Diklat : T. LN. 2008.11

No Buku : 11
Nama : Ari Handyani
Nama Pembimbing :
Judul Buku : Effect Of Grouping and Misspecification on Estimation of Income
Distribution
Absrak :
This research project develops a simulation procedure to assess the relative
performance of the Singh-Maddala, Dagum and beta-2 income distribution models in
terms of distribution misspecification and the effect of grouping on the Gini coefficient
and the headcount ratio. Under the “true” distribution, we simulate the data for individual
household income, assign them into the 40, 20, 10 and 5 income classes of equal size and
fit the three assumed modles to each data set. The distributional parameters which are
estimated by the maximum likelihood method are then used to calculate the Gini
coefficient as well as the headcount ratio. Next, the bias and the root mean squared error
for oth measures are calculated to quantify both grouping and model misspecification
effects.
The conclusion has been reached that the 10 groups is the minimal number of income
classes that could represent the income distribution reasonably well. The Singh Maddala
and the beta-2 models appear to be comparable, either when they are correctly specified
or when they are miss-specified for each other. On the other hand, the Dagum model
behaves totally different by only performing well it when it was correctly specified. As
the incorrect model specification, the Singh-Maddala model performs best for all group
sizes and for predicting both the Gini and headcount index, regardless of whether the true
distribution is the Dagum or the beta-2 distribution.

Diklat : T.LN 2008.12

No Buku : 12
Nama : Raden Chandra Ningrum
Nama Pembimbing :
Judul Buku : Impacts of crude Palm Oil (CPO) Export Tax Policy On
Economic Welfare in Indonesia
Absrak :
Rhis study evaluates the impacts of Indonesia Crude Palm Oil (CPO) export tax on
economic welfare by formulating an econometric model of the CPO industry in Indonesia
for the period April 1998 to Desember 2005. The welfare effects of export tax cover
distribution effects on CPO domestic price. Domestic supply, domestic and foreign
demand and also farmer price. The simulation, based on average data for the period
September 2000 to December 2006, was used to identifly the impacts of different export
tax rates. The results show that an increase of CPO export tax rate has caused a decrease
in CPO domestic price as well as farmer price, CPO domestic production and foreign
demand. On the other hand, the imposition of CPO export tax has positive effect on
domestic demand and results in net welfare gain to Indonesia. The gain of consumers is
relatively less than the loss of producers. Determining export tax rate at appropriate level
is necessary, because the higher the export tax, the lower the tax revenue and net gain to
Indonesia welfare. Futhermore, export tax policy could be an effective instrument to
guarantee an adequate supply in domestic market. It may also help to lower the CPO
domestic price, although the effect is not considerable and probably inadequate to
stabilize the cooking oil price.

Diklat : T.LN 2008.13

No Buku : 13
Nama : Heru Kusharjanto
Nama Pembimbing : Eiji Mangyo
Donghun Kim
Judul Buku : Infrastructure And Human Development : The Case Of Java
Absrak :
This study examines the between infrastructure and human development in jaya,
Indonesia. Using a panel data for the period 2002-2005, the paper analyzes the effects of
the infrastructures availability : electricity, clean water supply, road network, and the
number of classroom in senior high school on human development index (HDI). The
results show that all variables of infrastructures do have significant positive correlations
with the human development index and each component of HDI (Life expectancy at
birth, literacy rate, mean years of schooling, and real per capita expenditure).

Keywords : infrastructure, human development index, java, panel data.

Diklat : T. LN. 2008.14

No Buku : 14
Nama : Agus Hartanto
Nama Pembimbing : M. Zwanenburg, MA
Ir. Bakti Setiawan, MA, PHD
Judul Buku : Factor effecting the role of community based organization in
urban poverty projects ( A Comparison between Tanjung Hilir Kelurahan and Dalam
Bugis Kelurahan, Pontianak Municipality, Kalimantan barat Province)
Absrak :
Understanding performance and role local organization within community surround it is a
key for interventions from external actors (both government and nongovernmental). For
that reason, this research tried to identify factors that influence the performance of
community based organization in urban poverty project, particularly from internal factor
within community. Hence. The factors that were observed were : Social capita;
dimension community perception toward the performance of the BKM
This study began with the hypothesis that community based organization with higher
value dimension of social capital and higher value of internal factors will has higher
performance and role within society. Thus, the aim of the research is to describe, compare
and analyse the performance and thr role of BKM as a community based organizations
and factors that effect the performance and the role BKM. These aims are breaked down
into two research question. Those are : “what kind of factors do influence the
performance of BKM?” and to what extent do these factors influence the role of BKM?”.
To answer those questions, this study combined quantitative and qualitative method for
exploring the level of community participation in two kelurahans, which received the
program : Dalam Bugis kelurahan and Tanjung Hilir kelurahan. The quantitative method
was used by applying some indicators to measure their influence towards the role of
community-based organization from the community perception. Those indicators were :
social capital dimension such as : group characteristics, trust, togetherness and
volunteerism, and the internal factors that included: leaderships, internal democracy and
transparency. The qualitative method was used to analyze the reason or perception from
key persons toward some issues. The data collecting is carried out through
questionnaires, interviews and secondary data.
The main findings of the research identified that each indicator has a correlation with thr
outcome perception towards the role and the performance of BKM as a community base
organization. Mostly of these indicators has a positive correlation with the outcome
perception of the community base organization. Only indicator volunteerism has a
negative correlation. However, if the research employees all these indicators become one
equation in a linier regression with dependent variable is outcome perception, only
indicators leadership, internal democracy and togetherness have statistically significance
in influence the role of this organization.

Keywords : social capital, community base organization, internal factors, community

perception.

DIKLAT : T.LN.08.15
NO : 15
NAMA : Thomas Wunang Tjahjo
NAMA PEMBIMBING: Ir. Kawik Sugiana,M.Eng PHD
 Carley Pennink, MSc.
JUDUL BUKU : RISK ALLOCATION IN PUBLIC-PRIVATE
PARTNERSHIP (Case study of Giwangan Bus Terminal in Yogyakarta City)
ABSTRAK
A public Private Partnership (PPP) has to counter each risk that can disturb its project,
and potentially harm every party who involve in the PPP. Therefore, a PPP should know
every potential risk that they face or may face and mitigate each risk that may occur
during the project. How they recognize the risks and mitigate them should start from the
beginning of a project. PPP often fail because risks have not been appropriately taken into
account. This should be done also in Giwangan Bus Terminal, Yogyakarta City.
Indonesia that established in a PPP scheme. There are two kind of example that shows
risks for this partnership :
1.Demonstration from sellers on its opening ceremony;
2.Earthquake on May 27,2006 in Yogyakarta City that impacted on the terminal.
It is questionable whether Giwangan Bus Terminal partnership took potential risks into
account and managed them in PPP means.
This research assesses risks allocation and mitigation in Giwangan Bus Terminal
partnership. The research type is descriptive case study research. Main research question
in this research is how were risks in Terminal Giwangan PPP allocated and mitigated.
Giwangan Bus Terminal partnership is a partnership between a local government that is
Yogyakarta City Government and a private company which is Perwita Karya,Co.Ltd. For
both parties, the partnership is a novel project in PPP. This experience reflects in their
design of PPP, as well as the way they identifty, mitigate and allocate the risks.
The project design in this research consists consists of two main aspects: the bus terminal
as the infrastructure project and PPP model. The terminal is type A terminal which in
Indonesia is defined as bus terminal that serves public transport across provinces, across
cities, across rural areas, and within the city (Directorate General of Land Transportation
Indonesia 1995). The design of the model is in (Build-Operate-Transfer) BOT scheme
but the implementation is different which showed in its characteristics.
The characteristics of BOT in world Bank (1997b) in comparison with Giwangan Bus
Terminal partnership are:
 Asset ownership that in both World Bank model and Giwangan Bus Terminal
partnership belong to public and private
 Operation and maintenance that are done by public and private sector both World
Bank model and Giwangan Bus Terminal partnership belong to public and private
 Capital investment that provided by private sector in World Bank definition while
in Giwangan Bus Terminal partnership provided by private sector using public
sector asset as a guarantee to a bank.
 Commercial risk that is borne by private sector in World Bank definition while in
Giwangan Bus Terminal partnership is borne by both parties.
 Duration of contract which is 20-30 years in World Bank concept and 30 years in
Giwangan Bus Terminal contract.
Risk management needs process of risk iteration (Akintoye, Beck et al. 2003).
Risk iteration goes through the process of identification, mitigation, monitoring and
controlling of risks. Risks are also changing over time (project phase) and over market
(Lindfield 1998). Giwangan Bus Terminal partnership has the risks that exist in
correspond phase and market, the exception is in land market where the government has
released the land for the partnership. However, there is absence on the government side to
make precontract risk analysis (World Bank 1997a) since Yogyakarta City Government
wants all risks to be borne by private sector in this partnership. On the other side, there is
an indication of ‘the phenomenon of rent seeking behaviour on the part of a private
company with some monopoly power and the possibility of regulator capture’ that
pointed up by Public Choice Theory in this partnership (Lindfield 1998).
In the absence of motivation to identify risk, this partnership:
- has identified some risks but not in a purpose to identify risks
- ignored some critical suggestion from the consultation of experts.
Risks are allocated in Giwangan Bus Terminal partnership in the principle of all
commercial risks should be borne by the private sector, and not the party who is best able
to manage it (European Commission: Directorate-General Regional Policy 2003).
Giwangan Bus Terminal partnership is not a BOT project that tries to place all risks to
private partner, but still allocates some critical risks to the public sector.

Keywords: public private partnership, risk management, risk allocation, risk mitigation,
Build-Operate Transfer

Diklat : T. LN. 2008.16

No Buku : 16
Nama : Diah Ikawati
Nama Pembimbing :
Judul Buku : The Impact Of Economic Growth On Poverty Reduction in
Lampung Provice
Absrak :
Lampung is a province which has a large of agricultural production in Indonesia and
indeed in the world. However, this has apparently not brought any economic prosperty to
the Lampung society. More Surrisingly, Lampung has faced a serious challenge as the 8 th
poorest province in Indonesia. In line with these facts, this study has attempted to analyse
the changes in poverty and inequality from 1996 to 2006, the years ecompassing the
economic crisis. Data used in this study are taken from the consumtion data of Lampung
Province in 1996, 1999, 2002, and 2006 collected at household surveys, as parts of the
National Socio-economic Survey (SUSENAS) conducted by the Indonesia Central Board
of Statistics (BPS). Two levels of analysis, namely, bivariete and multivariate analysis are
applied to analysis the impact of economic growth and its redistribution to the poverty
reduction.

The bivariate analysis found that the economic of Lampung depends too much on the
agricultural sector. This was very unfortunate, because the economic crisis of mid 1997
compounded with the ET Nino disaster has swept away the province’s good performance
in the agricultural sector, which has indirectly led to an increase in the poverty rate after
the economic crisis. From other studies on Lampung, it has been found that the number of
people who fell below the poverty line in 1999, after the economic crisis, continued to
increase in both the urban and areas. Moreaver, the poverty gap index and poverty
severity index show that the condition in the rural areas was worse than that in the urban
areas. Similar patterns were found to be true event three years after the erisis. Even
though there was an improvement in the economic condition of Lampung society, as
indicated by the decreasing number of people falling below the poverty line, the
condition in the rural areas became worse than in the urban areas. In contrast, the increase
of fuel prices in 2005 contributed to worsening conditions of the urban society in 2006.
The poverty gap index and poverty severity index in the urban areas were much higher
than those in the rural areas.
Futhermore, the multivariate analysis using a regression model decopose the impacts of
economic growth and its redistribution to poverty reduction shows that the patterns of
poverty reduction in Lampung have fluctuated over the period 1999-2006. In the first
there year period (1996-1999), both urban and rural areas have experienced the impact of
economic crisis. However, the impact was worse in rural areas than in urban areas.
Furthermore, in the period 1999-2002, Lampung has successfully improved itseconomic
condition, with a decrease in the poverty rate, in both the urban and rural areas. However,
the condition in rural areas was still the same as in the previous three year period.
Interstingly, the opposite direction happened in the next four year period (2002-2006).
The rural areas were more economically developed than the urban areas and the
redistribution of that growth was more uniform in the urban areas.
The results presented in this study highlight the critical need to set up a short term and
medium term poverty alleviation strategy. In the short term strategy, the Government of
lampung could continue giving micro credit or subsidy to the poor people. However,
there should be a clear strategic planning and a good coordination between one program
to another to avoid misallocation budget and wrong targeting. Moreover, the economic
growth should be increased as well as keeping that growth to be distributed uniformly in
urban and rural areas. However, the poverty alleviation program could not be sustained
without accompanying by medium term strategy. The policies need to be adjusted to be
pro-poor to reduce the gap between the rich and the poor, such as providing the poor with
employment opportunities, redistributing land for the landless or near landless people,
giving to improve the society skill in agriculture, introducing appropriate agricultural
technology, and improving access to credit and infrastructure so they can purchase
agriculture inputs and get more profit. The policies also need to promote the small and
medium enterprises and support the service sectors that have close links with progress in
industry and agriculture. Moreover, the government of Lampung should also guarantee a
minimum wage according to the local conditions in order to make sure that the poor
workers will have adequate income to afford their basic needs. If the Government of
Lampung Province implements these suggestions, it would not be impossible for
Lampung to get out of the poverty trap in the future.

Diklat : T.LN.2009.17
No Buku : 17
Nama : Suswandi
Nama Pembimbing : Ryuta Ray Kato
Nawalage S. Cooray
Judul Buku : Bank Lending Channel of Transmission Mechanism : A Case of
Indonesia
Absrak :
The aim the research paper is to examine the existence of bank lending channel in
Indonesia, Bank Lending channel is based on analysis that banks have specific role in
financial system because bank can solve asymmetric information problem in credit
market. Since banks have specific role in financial system, banks can play vital role
transmission mechanism of monetary policy. We apply the quarterly balance sheet panel
data (2005ÏII-2008:III) on 64 commercial banks in Indonesia and to test effect of
monetary policy to the bank loans. Bank lending channel exists if central bank can effect
supply of bank loans. Bank that have special characteristics like size, liquidity and
capitalization will respond differently to the monetary policy shock. The estimation result
shows that bank lending channel in Indonesia exists. Special characteristics of bank like
size and capitalization are important factor in transmission mechanism of the monetary
policy while bank liquidity is not. The larger and well capitalization banks can reduce
negative effect of the monetary policy shocks than smaller and poor capitalization banks,
but capitalization is more important than size of bank. This result is consistent with
hypothesis of bank lending channel.

Keywords : Monetary policy, Bank Lending Channel Indonesia.

Diklat : T.Ln.2009.18
No Buku : 18
Nama : Idyah Fitriandari
Nama Pembimbing : Dr. Yonosuke Hara
Judul Buku : Comparing Technical Efficiency of Transjakarta’s Corridors
towards Sustainable Transportation.
Absrak :
The paper aimed to develop proper efficiency measurements for the TransJakarta busway
System in order to make recommendations for future improvements. Data Envelopment
Analysis (DEA) was used because it allowed for multiple input and outputs, and did not
require strong a priori assumptions regarding production technology. Total monthly
distance covered by fleets in km, passengers per km and average mileage covered by bus
per month were found to be proper efficiency measurements. Corridor 4 and 5 were
found to be robustly efficient with technical efficiency scores of 100% for most of the
conditions; corridor 1,3, and 6 were inefficient due to scale inefficiency; corridor 7 was
also inefficient due to pure technical inefficiency; and corridor 2 suffered in both. This
paper came to the conclusion that the scale inefficient corridors should convert a single
bus into an articulated bus; make a fully dedicated bus lane, and be integrated with
existing buses, while technical inefficient corridors should upgrade locak operations
though technology.

Keywords : TransJakarta, busway performance, DEA, technical efficiency analysis

Diklat : T.LN.2009.19
No Buku : 19
Nama : Ratna Setyowati
Nama Pembimbing : Dr. Katsuki Takao
Drs. Arief Rosyidie, MSP, M.Arch, PH.D
Judul Buku : Cuntinuity Planning Strategies In Tourism Industry
Absrak :
Tourism industries lie on condition which vulnerable to the risks from varied
sources. The risks can disturb the performance of tourism industry. In term of
organization. The risks can make timing and financial loss, human victim, decrease the
image or reputation, un-normally operational organization, etc.
The purpose of this study is to conduct an exploratory analysis of the risk in torism
industry. The specific research objective is to identify and determine the nature of risks
and the continuity strategies in the tourism industry. The research conducted in gathering
qualitative data using a questionnaire and interview.
Identification of risk factor is the first finding in this study and the internal risk factor
is annually faced by tourism industry rather than external risk factor. Most of tourism
industry’s response to handle the risk is “modify the process”. It caused by this response
is more flexible due to some limitations and conditions of tourism industry. Some
strategies have been conducted for continuing the operational of industry ever not on
perfectly yet. Such as the implementation of risk assessment just for the short term, not
sufficient allocation budget for maintaining the risks, the continuity plan is not well
documented, etc. It caused by some limitations and conditions in tourism industry. Even
though the implementation of BCP not in perfect process yet, also time and cost
consuming, there is significant responses from tourism industry towards the existing of
BCP. The implementation of BCP gives the benefit for the organizations. It reduce the
risks, respond over the event happened, recover from the impact and restore to the normal
operation.
To success the implementation of BCP need to consider the participatory manner. It
means that all parts of industry involved in the process and each part should understand
the nature and purpose of BCP. BCP process is not an easy task, but it can implemented
successfully when it bicomes a collective endeavour and not just one person lonely task.

Key words : Business continuity planning, internal and external risk factors, risk
assessment.
Diklat : T.LN. 2009.20
No Buku : 20
Nama : Ririn Hidayati
Nama Pembimbing : Dr. Masako Kurosawa
Dr. Yonosuke Hara
Judul Buku : Gender Earning Gaps : A Case of West Java Province, Indonesia
Absrak :
This paper studies the gender wage differential in urban and rural areas of West Java
Province, Indonesia. In order to investigate the main factors that leas to gender earning
gaps in West Java, a sample of 145,310 individuals aged 15 years and over who were
either working as employees or not working, drwn from cross sectional data of the
National Survey of Social Economic Issues (SUSENAS) for the period 2001 – 2005, is
used to estimate wage and accupation equation. The results suggest that working women
earn less than men in all sectors and occupations, and are concentrated in lower paying
occupations. In particular, decompositions reveal that the gender earnings gaps are
largely due to unexplained factors, which are 59.20 percent in urban area and 66.43
percent in rural area. This result does not rule out the possibility of discrimination in
terms of gender segregation.

Key words : Employment, wage differentials; occupation; discrimination

Diklat : T.LN.09.21
No : 21
Nama : Peri Gandara
Nama pembimbing : Ir.Tubagus Furqon Sofhani,MA.,Ph.D
DR. Hiroyuki Sakakibara
Judul:Identification Of Conflicts In Greater Bandung Area For Implementing Sustainable
Development
Absrak
There are some conlicts on environments in Greater Bandung Area. Such conflict
occur not only between government and NGO but also among governmental institutions.
In this study, the conflicts which occurred in Greater bandung Area analyzed. Policy
conflict model is applied to these conflicts. Based on the results of analysis. Some
suggestions for managing conflicts are made.
Basic idea of “policy Conflict Model” is game theory but in game theory, outcomes
are represented by combination of strategies. However, because it is difficult to define
strategies, alternatively model is constructed based on concerns (Y). The purpose of game
theory is foresasting player’s behaviors while this model is applied to find the way to
achieve cooperation. The component of “Policy Conflict Model” is participants concerns
and preferences. Using sets of concerns, the relationship between two participants can be
classified into the following six types. Based on the typology relationship, group of
participants can be recognized. The purpose of grouping is to recognize the principal
conflict and the potential mediator.
Conlict on waste management is extremely complex so we decided to be separated 3
phase. The conflict appears openly after landslide in Leuwigajah disposal site. Provincial
government persuades Bandung Regency and the other cities to open TPA (Final disposal
site) and to accept garbage from Bandung Municipality. To solve waste problem,
bandung Municipality plan to contruct incinerator plant in Gedebage but project was
protested by local residents. When incinerator is not yet constructed, a capacity of
fisposal sites in Greater Bandung Area was overload. Provincial government wants to
coordinate municipalities and regencies on solid waste management, to reopen
leuwigajah site. Leuwigajah residents which supported by NGO indeed reject this
planning.
In phase 1, it is a conflict between municipalities and suburban area. Expert and
Provincial Government can play roles as mediators. In phase 2, It is conflict between
local residents and municipalities. Expert and Environmentalists is potential mediators.
And in phase 3, it is a conflict between NGO and government. Local governments can
play roles as mediators. Based on these formulations some analysis for managing
conflicts and suggestions for solving problem are made.

Key Word : Players, strategy, conflict, participants, concerns, preferences, mediator

Diklat : T.LN.2009.22
No : 22
Nama : Widyo Pura Buana
Nama pembimbing : Ir.Haryo Winarso,M.Eng.,PhD
Judul : Vulnerability of Communities Living in and near Riwo
Forest,Indonesia: Towards Sustainable Livelihoods
Abstrak
The research focuses on the economic use of the forest ecology by ‘Masyarakat Desa’
communities living in Riwo forest Dompu Regency and the adverse forestry policy of the
Indonesian government. The forest supports livelihood of people who inhabit inside or in
the vicinity. At the same time, however, it is very valuable resources for a central and
local government to earn money from outside through exports. It also has functions to
control natural disaster such as flood and landslide. In addition, it currently seems to be
one of the most valuable resources to protect global warming. Thus, the utilizations and
reservations of the forest resources are always laid under the careful consideration of
governmental economic and environmental policies, which do not always benefit the
forest people, or rather give negative influences on them, resulting in very vulnerable
condition to people’s livelihoods. Most of the people are poor and heavily dependent on
forest products like wood, fuel wood, honey and paddy cultivation. The government
should enact a forest policy where the communities are able to use the forest sustaining
and yet be able to renew sustainability ecology, especially to allow people to cultivate in
buffer zone of Riwo forest.

Keywords : Community forestry, Vulnerability, Common Property Resources, Dompu,

Indonesia.

Diklat : T.LN.2009.23
No : 23
Nama : Ouceu Satyadipura
Nama pembimbing :
Judul Buku : Female Labour force participation in Papua province, Indonesia:
Factors affecting the pattern and its share on household poverty reduction
Abstrak
FEMALE labour force participation rates in Indonesia in the last 30 years have
increased sharply in modern sectors and skilled jobs. However women’s participation in
the labour force is still considered as one of the most critical challenges for the
development in Indonesia. Most of the women have weak labour market position and low
salary, which influence the household income and household poverty level. In contrast,
the conventional relationship between women’s participation in the labour force and
poverty appears to be inversely related meaning that higher labour force participation by
women should lead to a reduction in the level of poverty. The purposes in this study were
to identify the factors influencing female labour force participation, and to observe the
relationship between female labour force participation rates and household poverty rates
in Papua province, Indonesia 2007.
This study used the National Socioeconomic Survey of Indonesia (SUSENAS) 2007
dataset as main data. The unit for analysis of this study was women aged 15-64 years in
the labour force in Papua province, Indonesia 2007, based on a sample of 6,415
respondents of women aged 15-64 years. This study also used a descriptive analysis to
measure the central tendency and dispersion on the data, bivariate analysis such as cross
tabulation and chi-square to measure the inferences of population means of the data, and
a binary logistic regression as multivariate analyses to measure, explain and predict the
degree of relationship among the variates.
The results of this study revealed that demographic, socio-cultural and economic
factors have a significant influence on women’s employment in Papua province. In terms
of demographic factors , the bivariate analysis suggests that age of the woman has the
highest association to employment compared to her marital status and number of
children. A strong relation was also observed for women’s education and their place of
residence on women’s employment. Household headed types and household expenditure
factors also have an important role on influencing women’s decision to join the labour
force. By using a multivariate analysis, such demographic, socio-cultural and economic
variables have shown a strong significant influence on female labour force participation
in Papua province. The findings of the multivariate analysis result suggested that those
women aged 15-64 in Papua province, who are older, who have fewer children, who are
ever married, who have lower education levels, who live in rural areas, who come from
women-headed family type, and come from family which live below poverty line, share a
higher probability to join the labour force than others. In terms of poverty thirty seven
percent of women in Papua province live in the families which cannot provide to 2,100
kilo calories per person per day. By using logistic regression coefficients, it is predicted
that women aged 15-64, who have higher education, who live in urban areas, who come
from small family size, and who have a job, share a higher probability to avoid poverty.
Moreover, by using an Indonesian national poverty line for Papua province (2,100 kilo
calories or U$$ 0.75 per-day per-person), the analysis also suggests that women’s
employment have a significant effect on household poverty reduction either for couple-
headed households or for woman-headed households. The Indonesia national poverty line
for Papua province 2007 (U$$ 0.75 per-day per-person) is based on the ability of each
person to afford the minimum necessity of 2,100 kilo calories per person per day.
The author recommends that to strengthen the national or regional policies to
encourage women to participate in higher education level, since more than 30 percent of
women aged 15-64 never attended school and at the same time they have the highest
labour force participation rate with very low wage rates. The policy can be implemented
by increasing the wage level of workers and strongest the free basic eduation cost policy.
It is recommended that the government must focus on rural areas as the first priority of
economic development and increase the minimum wage for women, so that the poverty
of women in Papua would reduce. In terms of poverty reduction, a pro-woman labour
policy must be introduced to encourage greater women’s participation.

Diklat : T.LN.2009.24
No : 24
Nama : Mrs.Nucke Widowati Kusumo Projo
Nama pembimbing :
Judul Buku : Impact of Health Insurance For the Poor (ASKESKIN) On Health
Services Utilization In Indonesia
Abstrak
Health insurance for the poor (ASKESKIN) is insurance program to protect the
poor. ASKESKIN provides free services mainly in public health facilities, based on the
referral system ranging from outpatient care in the community health center to tertiary
health care services for inpatients in the district hospital. The main purpose of this
program is to increase health utilization and quality of care among the poor in Indonesia.
The aims of this study are to investigate the distribution of ASKESKIN program among
the poor, is effect on the health service utilization and the pattern of health care choice.
Data in this study was obtained from Indonesia’s Socio Economic Survey
(SUSENAS) Health and Housing Module 2004 and 2007. However, the samples used in
the study include only persons aged 19-60 years in the lowest income quintile to
represent the poor. Besides the descriptive analysis for the ASKESKIN distribution, the
logistic regression, and the OLS regression model are used to analyze the effect of
ASKESKIN on health care utilization. The multinomial logit model is used to study the
pattern of health care choice under ASKESKIN scheme. In addition to ASKESKIN
enrollment variable, this study also includes other variables that determine health service
utilization among the poor i.e. age, sex, marital status, self reported illness, day with
illness years of schooling , family size, location, income, year of insurance, health
worker ratio, health facility ratio, distance and time to the nearest health facilities.
The results of the study show that distribution of ASKESKIN has effected overall
outpatient utilization, overall inpatient utilization, public outpatient, public inpatient,
and private inpatient utilization. Nevertheless, the ASKESKIN program seems to be
insignificantly to private outpatient utilization. ASKESKIN program has the positive
effect on the pattern of health care choice. It means that people under ASKESKIN
program are more likely to choose public health facility rather than traditional healers of
self treatment. Policy reconmmendations for improving the ASKESKIN program are as
follows : (1) Improving the distribution of ASKESKIN enrollees to avoid the leakage of
the program to the non poor; and (2) for the increase in health care utilization, the
government should improve the health care referral system to reduce the cost.

Diklat : T.LN.2010.25
No : 25
Nama : IMAM MACHDI
Nama pembimbing :
Judul Buku : A Study on Parallel Holistic Twig Joins for XML Query
Processing
Abstrak
Along with the growing popularity of the World Wide Web,XML query processing
systems play such important roles to process query patterns against such large XML
database. One of the core operations in XML query processing system is to find in XML
database all matches that satisfy a specified query pattern. Among other algorithms for
processing query patterns, the family of holistic twig join algorithms is prominent due to
the performance advantage and the capability of handling complex query patterns.
Several crucial issues have risen on XML query processing systems that typically
manage large collections of heterogeneous XML documents and processes numerous
concurrent query patterns. By the time, the tremendous size of XML documents and
query patterns definitely overwhelms the processing capabilities to maintain good
performance. This challenging issue to maintain and improve the system performance has
attracted many researches in the area of parallel XML query processing on shared-
nothing cluster system and multi-core system. Another challenging issue-XML data
partitioning, which is fundamental to provide the basis of data parallelism, is essentially
complex to perform since XML data and its structure have to be preserved in partitions.
In addition to data parallelism, other forms of parallelism such as task parallelism, intra-
query parallelism and inter-query parallelism are considered as an important issue to
potentially improve the parallel XML query processing system performance. Hence, we
see the challenging issues as the problem addressed in the parallel holistic twig join for
XML query processing.
This dissertation compiles the study on parallel holistic twig joins for XML query
processing and presents three main contributions, First, we propose a novel static XML
data partitioning scheme on a cluster system, the Grid Metadata for XML (GMX). GMX
is a conceptual model that is composed of XML metadata derived from static information
of XML documents and queries in the past. The model specifies relationship between
XML documents and queries in the form of grids as logical partitions. Based on the
relationship, a grid is associated with a cost of processing queries for XML documents.
The basic notion of partitioning is to decompose a grid with a high cost to finer sub-grids
with lower costs. To partition XML data logically in grids, GMX is facilitated with a set
of hierarchical partitioning steps: XML document clustering, query clustering, document-
based refinement, query-based refinement, and subquery-based refinement that yield
partitions from coarse granularity to relatively fine granularity, respectively. Different
partition granularities will assist the allocation approach to allocate partitions for
achieving workload balance in a cluster system.
The second contribution is to propose the Stream-based Partitioning for XML (SPX),
which is an on-the-fly XML data partitioning scheme, on a cluster system.
The SPX is aimed for handling workload imbalance during query execution due to
changing patterns of query execution in the system. It repartitions XML data, which is
previously partitioned and allocated by GMX scheme, and reallocates partitions from
highly loaded processing nodes to lightly loaded processing nodes for achieving dynamic
workload balance. The target of partitioning is streams of XML nodes, rather than XML
documents. The SPX scheme characterizes itself as straightforward computation and
providing finer partitions, which are certainly appropriate for on-the-fly partitioning.
Besides, a partition generated by the SPX scheme contains complete solutions for a query
without data dependency on other partitions and eliminates unnecessary XML nodes for
faster processing.
Lastly, we propose a framework of the parallel holistic twig join algorithm on a
multi-core system as the third contribution. The framework aims at developing general
parallelism techniques for data parallelism and task parallelism to speed up a single long
running query. In data parallelism, the SPX scheme is adopted to partition streams of
XML on-the-fly. An important technique is to estimate the partition size for providing
finer parallelism and reducing memory access contention. In task parallelism, the holistic
twig join algorithm is decomposed into two tasks that are parallelized using the pipelining
technique. Another important technique is to estimate the size of data transfer from task
one to task two for achieving optimal parallelism. As the result, the performance of task
parallelism is able to enhance the performance of data parallelism.

Diklat :
No Buku :
Nama :
Nama Pembimbing :
Judul Buku :
Absrak :

Diklat :
No Buku :
Nama :
Nama Pembimbing :
Judul Buku :
Absrak :

Diklat : T.TL.2005.27
No : 27
Nama : Alex Stuckey
Nama pembimbing : Dr.Craig H.McLaren
Judul Buku : Time Series Analysis for Official Government
Agencies
Abstrak
This course will consider issues which impact on time series analysis of
official government statistics. Basic theory of seasonal adjustment concepts
will be considered with topics including estimating seasonality, prior
corrections, and trend estimation. Advanced topics covered will include
modelling and choosing appropriate univariate ARIMA models, correcting
for moving holidays, trading day, outlier adjustments, revisions, aggregation
issues, estimating changes in the seasonal pattern, and estimating changes in
the level. Practical exercises, including computer based examples, will be
used.

Diklat : T.TL.2005.28
No : 28
Nama : Ninik Anisah
Nama pembimbing : Professor DR.Eng.Masaaki TAGURI
Judul Buku : Application of Bootstrap Method to Imputation in
Survey Data
Abstrak
One problem in survey sampling is that unit or item non response occur
frequently. Compensating to this problem, imputation methods are often
employed using auxiliary data. Not only deterministic imputation, such as
ratio and regression imputation, but also random imputation such as ratio
random and regression random imputation can be used. However, treating
the imputed data set and directly applying existing methods especially in
bootstrap method, it does not produce the valid result because bootstrap
method does not account for the effect of missing data. The application of
bootstrap method to the variance estimation and statistical inference, which
will give asymptotically valid result, should take into consideration sampling
design, imputation methods and type of statistics used in inference. If there
are missing data the use of naive bootstrap estimator makes underestimate,
unless the imputation of bootstrap data set is done in the same way as
original data sets. The objective of this article is the comparison of the
imputation methods make use of auxiliary data observed for all data. The
viewpoints for the comparison are relative bias, variance, efficiency and
distribution function. As for the imputation methods, we use ratio and
regression methods for deterministic imputation and ratio and regression
random methods for random imputation and the considered parameter is the
population total. From simulation results, the relative bias and the variance
increase as the response rate and the correlation coefficient decrease. The
deterministic imputation method dependent on the condition of the auxiliary
variable, when the auxiliary variable is not equal to 1 then the deterministic
imputation is the best imputation method. Although the random imputation
is not the best imputation method but independent on the condition of the
auxiliary variable therefore we may say that the random imputation can
handle in any situation. In random imputation we can say that regression
random imputation may be the best method to impute missing values.
This fact had proved that this method can reduce the variance compared to
ratio random imputation. The superiority of regression random imputation to
preserve the distribution function can be shown because the bias of
parameter.