Application of Geographically Weighted Regression for Vulnerable Area Mapping of Leptospirosis in Bantul District

Abstract Geographically Weighted Regression (GWR) is regression model that developed for data modeling with continuous respond variable and considering the spatial or location aspect. Leptospirosis case happened in some regions in Indonesia, including in Bantul District, Special Region of Yogyakarta. The purpose of this study are to determine local and global variable in making vulnerable area model of Leptospirosis disease, determine the best type of weighting function and make vulnerable area map of Leptospirosis. Alos satelite imagery as primary data to get settlement and paddy fields area. The others variable are the percentage of population’s age, flood risk, and the number of health facility that obtained from secondary data. Determinant variables that affect locally are flood risk, health facility, percentage of age 25-50 years old and the percentage of settlement area. Meanwhile, independent variable that affects globally is the percentage of paddy fields area. Vulnerability map of Leptospirosis disease resulted from the best GWR model which used weighting function Fixed Bisquare. There are 3 vulnerable area of Leptospirosis disease, high vulnerability area located in the middle of Bantul District, meanwhile the medium and low vulnerability area showed clustered pattern in the side of Bantul District. Abstrak Geographically Weighted Regression (GWR) adalah model regresi yang dikembangkan untuk memodelkan data dengan variabel respon yang bersifat kontinu dan mempertimbangkan aspek spasial atau lokasi. Kejadian Leptospirosis terjadi di beberapa wilayah di Indonesia termasuk di wilayah Kabupaten Bantul Daerah Istimewa Yogyakarta. Tujuan dari penelitian ini adalah menentukan variabel lokal dan global dalam membuat model kerentanan Leptospirosis dan menentukan jenis fungsi pembobot yang terbaik serta membuat peta kerentanan wilayah Leptospirosis menggunakan aplikasi GWR. Citra Satelit Alos digunakan untuk mendapatkan data penggunaan lahan, yang selanjutnya diturunkan menjadi prosentase luas permukiman dan sawah. Parameter lainya adalah prosentase umur penduduk, resiko banjir dan jumlah fasilitas kesehatan yang diperoleh dari data sekunder. Variabel yang berpengaruh secara lokal adalah Risiko Banjir, Fasilitas Kesehatan Presentase Usia 25-50 tahun, Prosentase Luas Pemukiman, sedangkan variabel independen yang bepengaruh secara global adalah Presentase Luas Sawah. Peta kerentanan Leptospirosis yang dihasilkan dari model GWR terbaik yaitu menggunakan fungsi pembobot Fixed Bisquare. Terdapat 3 kelas kerentanan Leptospirosis yaitu kelas kerentanan tinggi berada di desa-desa di tengah Kabupaten Bantul, sedangkan kelas sedang dan rendah menunjukkan pola menggelompok di wilayah pinggiran Kabupaten Bantul

Download Full-text

KELAYAKAN FINANSIAL USATANI KEDELAI

MIMBAR AGRIBISNIS Jurnal Pemikiran Masyarakat Ilmiah Berwawasan Agribisnis ◽

10.25157/ma.v5i2.2265 ◽

2019 ◽

Vol 5 (2) ◽

pp. 293

Author(s):

Dedi Djuliansah ◽

Trisna Insan Noor ◽

Yosini Deliana ◽

Meddy Rachmadi

Keyword(s):

Secondary Data ◽

Paddy Fields ◽

Primary Data ◽

Survey Method ◽

Production Volume ◽

Land Area ◽

C Value ◽

Break Even Point ◽

The Cost ◽

Planting Season

This study aims to identify the cost, revenue, and income of soybean farming, identify the feasibility of soybean farming, identify the breakeven point and change the break-even point due to changes in selling prices in Jatiwaras District, Tasikmalaya Regency. The method used in this study is a survey method, while the data used consists of primary data and secondary data. Determination of sample farmers using the Two Stage Cluster Random Sampling method, with a sample size of 65 people with a proportion of 27 farmers in paddy fields and 38 farmers in land, from a population of soybean farmers as many as 185 people. The results of this study indicate that the cost of soybean farming per hectare in paddy fields is Rp. 5,896,896.90 with receipts of Rp 8,478,139.53 and income of Rp. 2,581,242.63, while the cost of soybean farming per hectare on land is Rp. 4,163,487.48 with receipts of 8,342,774.57 and income of Rp. 4,179,287.09. Soybean farming in land is more feasible to be cultivated with an R / C value of 2.01 while the R / C value in paddy fields is 1.45. Minimum acceptance received by farmers from soybean farming so as not to lose in one planting season of Rp. 63,911.14 in paddy fields and Rp. 668,378.02 in land, the minimum production volume received by farmers from soybean farming so as not to lose in one planting season is 10.65 Kg in paddy fields and 111.40 Kg in land and minimum land area that must be processed by farmers so that no loss in one planting season of 0.01 ha in paddy fields and 0.08 ha on land. Decrease in output price of Rp. 1,000.00 (16.67%) causes the minimum acceptance received by farmers from soybean farming so as not to lose in one planting season of Rp. 100,196.38 in paddy fields and Rp. 767,384.61 on land. The margin value of safety on soybean farming is 90.53 in wetland and 82.40 in land area

Download Full-text

Estimation of Error Variance-Covariance Parameters Using Multivariate Geographically Weighted Regression Model

Abstract and Applied Analysis ◽

10.1155/2020/4657151 ◽

2020 ◽

Vol 2020 ◽

pp. 1-5 ◽

Cited By ~ 1

Author(s):

Sri Harini

Keyword(s):

Statistical Inference ◽

Geographically Weighted Regression ◽

Weighting Function ◽

Likelihood Estimation ◽

Error Variance ◽

Least Square ◽

Weighted Regression ◽

Inference Procedure ◽

Geographically Weighted Regression Model ◽

Estimated Error

The Multivariate Geographically Weighted Regression (MGWR) model is a development of the Geographically Weighted Regression (GWR) model that takes into account spatial heterogeneity and autocorrelation error factors that are localized at each observation location. The MGWR model is assumed to be an error vector ε that distributed as a multivariate normally with zero vector mean and variance-covariance matrix Σ at each location ui,vi, which Σ is sized qxq for samples at the i-location. In this study, the estimated error variance-covariance parameters is obtained from the MGWR model using Maximum Likelihood Estimation (MLE) and Weighted Least Square (WLS) methods. The selection of the WLS method is based on the weighting function measured from the standard deviation of the distance vector between one observation location and another observation location. This test uses a statistical inference procedure by reducing the MGWR model equation so that the estimated error variance-covariance parameters meet the characteristics of unbiased. This study also provides researchers with an understanding of statistical inference procedures.

Download Full-text

PEMODELAN INCIDENT RATE DEMAM BERDARAH DENGUE DI INDONESIA YANG BERKAITAN DENGAN FAKTOR LINGKUNGAN MENGGUNAKAN METODE GEOGRAPHICALLY WEIGHTED REGRESSION (GWR)

EKOLOGIA ◽

10.33751/ekologia.v20i2.2167 ◽

2020 ◽

Vol 20 (2) ◽

pp. 64-73

Author(s):

Kiki Amelia ◽

Latifa Oktafiani Asril ◽

Lasmi Febrianti

Keyword(s):

Population Density ◽

Dengue Hemorrhagic Fever ◽

Geographically Weighted Regression ◽

Physical Environment ◽

Secondary Data ◽

Hemorrhagic Fever ◽

Least Square ◽

Gaussian Kernel ◽

Weighted Regression ◽

Incident Rate

Dengue hemorrhagic fever cases in Indonesia often occur in cities and villages. Every year hundreds to thousands of people must be hospitalized due to this disease. There are several factors of the physical environment that directly or indirectly influence the transmission of this disease. Such as rainfall, air temperature, and humidity. In addition to the physical environment there are several other factors that can increase the occurrence of dengue cases, namely population density and the level of larvae free in an area. For this reason, we conducted a study of the above factors and their contribution in the addition of dengue cases that occurred in Indonesia in 2015 using secondary data. The purpose of this study is to identify and make a BDB iricident rate model related to environmental factors such as temperature, humidity, population density, and the amount of rainfall on the number of cases of dengue hemorrhagic fever in Indonesia in 2015. The method used is the Geographically Weighted Regression method. (GWR). In the GWR model the parameter estimation uses Weighted Least Square (WLS) by weighting the gaussian kernel function. The results of the study concluded that modeling with GWR was better than linear regression and the variables were significantly different in each region.

Download Full-text

PERSEPSI MASYARAKAT TERHADAP PENGGUNAAN LAHAN SAWAH DI DESA TOUNELET SATU KECAMATAN SONDER

AGRI-SOSIOEKONOMI ◽

10.35791/agrsosek.14.1.2018.19192 ◽

2018 ◽

Vol 14 (1) ◽

pp. 157

Author(s):

Lovi Cherry Valentine Katuuk ◽

Vicky R. B. Moniaga ◽

Juliana R. Mandei

Keyword(s):

Land Use ◽

Public Perception ◽

Sampling Method ◽

Secondary Data ◽

Rice Fields ◽

Paddy Fields ◽

Primary Data ◽

Research Report ◽

Online Media ◽

Pig Farms

This study aims to determine the Perceptions of the Community Against Rice Land Use In Tounelet Village One, Sub-district Sonder. This research was conducted from November 2017 until January 2018, starting from preparation until preparation of research report. The research site is Tounelet Village One, Sub-district Sonder. This research uses primary data and secondary data. Primary data was obtained from 30 respondents by using questionnaire, while secondary data was obtained from print media and online media and from journals in the literature related to this research. Sampling method in this research using purposive sampling method. Result of research indicate that public perception toward Wetland Field Estate is good with perception index 77.27%. The community agrees with the use of existing paddy fields in Tounelet One village, Pigs help both as rice fields and as pig farms.

Download Full-text

ANALISIS KONVERSI LAHAN SAWAH DI KOTA SOLOK

Agrifo Jurnal Agribisnis Universitas Malikussaleh ◽

10.29103/ag.v3i1.675 ◽

2018 ◽

Vol 3 (1) ◽

pp. 39

Author(s):

. Mardianto

Keyword(s):

Agricultural Land ◽

Secondary Data ◽

Simple Random Sampling ◽

Paddy Fields ◽

Primary Data ◽

Survey Method ◽

Small Scale ◽

Land Conversion ◽

Internal Factor ◽

Factors Affecting

Agricultural land conversion occurs mostly in big cities in Indonesia and also occurs in small villages and towns on a small scale but not much has been done by the study. This study was linked to detect factors affecting land conversion in Kota Solok. This research was conducted by survey method. Sampling is done by simple random sampling with balanced amount. The data collected in this study includes primary data and secondary data. The analysis is done by description using percentage of respondent's level of achievement (tcr). The result of the analysis shows that the conversion of paddy fields in Solok City is mostly done by individual buyers, the internal factor which has the greatest effect on the conversion of paddy fields in Solok City is the economic condition, while the external factor is caused by the population growth and the policy caused by the weakness of policy control which government apparatus.

Download Full-text

Flood Risk Assessment of Residential Neighbourhoods in Calabar Metropolis, Cross River State, Nigeria

Environment and Natural Resources Research ◽

10.5539/enrr.v6n2p115 ◽

2016 ◽

Vol 6 (2) ◽

pp. 115 ◽

Cited By ~ 1

Author(s):

Benedict E. Ojikpong ◽

Bassey E. Ekeng ◽

Ukpali E. Obonga ◽

Samuel I. Emri

Keyword(s):

Flood Risk ◽

Flood Control ◽

Residential Buildings ◽

Drainage System ◽

Secondary Data ◽

Least Square Method ◽

Least Square ◽

Control Measures ◽

Primary Data ◽

Residential Areas

<p class="1Body">The study is aimed at examining the vulnerability of some residential neighbourhoods in Calabar to the menace of flooding with a view to determining residential areas of high, medium and low flood risk. Two hypotheses were formulated such as: there is no significant relationship between the magnitude of flood, and the vulnerability of residential neighbourhoods and the elements-at-risk to flood in residential neighbourhoods in Calabar do not vary significantly according to the topography of the area. The major primary data were obtained from the metric measurement of the coverage of flood and the assessment of the numerical value of the residential buildings considered vulnerable to flood within the areas measured. Secondary data were also obtained from the collection of both published and unpublished materials and data on flooded buildings and displaced persons were also obtained from the State Emergency Management Agency (SEMA), Calabar. The data were analyzed using descriptive statistics and hypotheses tested using the regression coefficient of the least square method and scatter grams for prediction. The results of the hypotheses were found to be significant as the magnitude of flood determined the vulnerability of some residential neighbourhoods. Vulnerability was found to be higher in low lying residential neighbourhoods. The study, however, recommends among others, planned and autonomous adaptation responses, flood plain zoning to urban agriculture, landscaping and recreational uses. Proper channelization of Calabar urban drainage system, stringent flood control legislation, and development control measures should be enforced so as to discourage people from building on or near flood-prone areas of Calabar.</p>

Download Full-text

Direct modeling of subway ridership at the station level: a study based on mixed geographically weighted regression

Canadian Journal of Civil Engineering ◽

10.1139/cjce-2018-0727 ◽

2020 ◽

Vol 47 (5) ◽

pp. 534-545 ◽

Cited By ~ 2

Author(s):

Hongtai Yang ◽

Taorang Xu ◽

Dexin Chen ◽

Haipeng Yang ◽

Li Pu

Keyword(s):

Geographically Weighted Regression ◽

Weighted Regression ◽

Independent Variables ◽

Employment Density ◽

Direct Modeling ◽

Global Variable ◽

Spatially Varying ◽

Gwr Model ◽

Metro System ◽

Better Than

Station-level ridership modeling is one of the ways to forecast metro ridership and reveal how factors influence ridership. Previous studies assumed that the relationships between the dependent variable and independent variables are either global or local, as indicated by the global model or the geographically weighted regression (GWR) model. This study explores the possibility that some independent variables have spatially varying relationships with metro ridership while others have constant relationships by employing the mixed GWR model. Data from the Chicago metro system were used. To establish an effective forecasting model, possible influencing factors are collected. OLS model results indicate that the proportion of recreational jobs to total jobs, number of bus stops, employment density, number of high-income workers, and the type of station (transfer or terminal) are significant variables influencing station-level metro ridership. By using the mixed GWR model, we find that the proportion of recreational jobs to total jobs is a global variable while the others are local variables. By comparing the results of mixed GWR, full GWR, and OLS models, we find that mixed GWR fits the data better and the residuals are less correlated. However, results of cross-validation indicate that the prediction power of the OLS model is better than that of the full and mixed GWR models.

Download Full-text

A new spatial-attribute weighting function for geographically weighted regression

Canadian Journal of Forest Research ◽

10.1139/x05-295 ◽

2006 ◽

Vol 36 (4) ◽

pp. 996-1005 ◽

Cited By ~ 14

Author(s):

Haijin Shi ◽

Lianjun Zhang ◽

Jianguo Liu

Keyword(s):

Geographically Weighted Regression ◽

Weighting Function ◽

Weighted Regression ◽

Attribute Weighting ◽

Geographical Distances ◽

The Subject ◽

Model Residuals ◽

The One ◽

Gwr Model ◽

Geographical Space

In recent years, geographically weighted regression (GWR) has become popular for modeling spatial heterogeneity in a regression context. However, the current weighting function used in GWR only considers the geographical distances of trees in a stand, while the attributes (e.g., tree diameter) of the neighboring trees are totally ignored. In this study, we proposed a new weighting function that combines the "geographical space" and "attribute space" between the subject tree and its neighbors, such that (1) neighbors with greater geographical distances from the subject tree are assigned smaller weights, and (2) at a given geographical distance, neighboring trees with sizes that are similar to that of the subject tree are assigned larger weights. The results indicate that the GWR model with the new spatial-attribute weighting function performs better than the one with the spatial weighting function in terms of model residuals and predictions for different spatial patterns of tree locations.

Download Full-text

KELAYAKAN USAHATANI KEDELAI DI LAHAN DARAT DAN LAHAN SAWAH

MIMBAR AGRIBISNIS Jurnal Pemikiran Masyarakat Ilmiah Berwawasan Agribisnis ◽

10.25157/ma.v5i2.2390 ◽

2019 ◽

Vol 5 (2) ◽

pp. 346

Author(s):

Rian Kurnia ◽

Trisna Insan Noor ◽

Eliana Wulandari ◽

Meddy Rachmadi

Keyword(s):

Random Sampling ◽

Sampling Method ◽

Secondary Data ◽

Paddy Fields ◽

Primary Data ◽

Survey Method ◽

A Value ◽

Cluster Random Sampling

This study aims to determine the feasibility of Soybean farming in dryland and paddy fields land in the Jatiwaras Subdistrict, Tasikmalaya Regency. The method used in this study is the survey method, while the data used consists of primary data and secondary data. Determination of sample farmers using the multistage cluster random sampling method. The number of respondents was taken as many as 36 farmers who were divided by farmers on 21 farmers on dryland and 15 farmers on wetland. The results of this study indicate that soybean farming in dryland is more feasible to cultivate with a value of R/C 1.98 while the value of R / C in wetland is 1.62.

Download Full-text