PREDICTION AND VISUALIZATION OF RELATIVE RISK MAPS USING AUTOREGRESSIVE DISTRIBUTED LAG (ADL) AND CONDITIONALLY AUTOREGRESSIVE (CAR) MODELS

Dengue fever (DHF) is an infectious disease caused by one of four different dengue viruses and is transmitted by mosquitoes, mainly Aedes aegypti and Aedes albopictus, found in tropical and subtropical regions. Dengue fever (DHF) is still one of the major public health problems in Indonesia. Along w...

Full description

Saved in:
Bibliographic Details
Main Author: Qodri Alfairus, Muh.
Format: Theses
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/77214
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
id id-itb.:77214
spelling id-itb.:772142023-08-23T13:32:41ZPREDICTION AND VISUALIZATION OF RELATIVE RISK MAPS USING AUTOREGRESSIVE DISTRIBUTED LAG (ADL) AND CONDITIONALLY AUTOREGRESSIVE (CAR) MODELS Qodri Alfairus, Muh. Indonesia Theses DHF, autorgressive distributed lag (ADL), conditionally autoregressive (CAR), Semarang city. INSTITUT TEKNOLOGI BANDUNG https://digilib.itb.ac.id/gdl/view/77214 Dengue fever (DHF) is an infectious disease caused by one of four different dengue viruses and is transmitted by mosquitoes, mainly Aedes aegypti and Aedes albopictus, found in tropical and subtropical regions. Dengue fever (DHF) is still one of the major public health problems in Indonesia. Along with increasing mobility and population density, the number of sufferers and the area of spread is increasing. This can be caused by climate change and low awareness to maintain environmental hygiene. This can be anticipated by modeling cases with the Autorgressive Distributed Lag (ADL) model and predicting for each region to map the risk into relative risk groups using the Localized Conditionally Autoregressive (CAR) Model. In this discussion, ADL and CAR models are used to see how far the risk of dengue cases can be formed and can anticipate the riskiest locations first. This research focuses on examining dengue cases in Semarang City from 2016 to 2021. The methodology begins with checking stationarity by looking at the P-value that is smaller than alpha, which is 0.05. The second step is to determine the appropriate model by looking at the cointegrity test and also checking the best model of the corresponding ADL. The third stage is modeling the ADL that is in accordance with the best model ADL (1,12) with ????2= 68% and making predictions for each sub-district in Semarang City. The next stage determines the value of the Moran Index to see whether autocorrelation occurs or not. Next, the best CAR modeling is used to map the risk based on the selected group. In this case, two clusters were obtained with a WAIC value of 108.43. These results indicate that the riskiest locations are divided into two, namely: Group 1 with low relative risk of DHF cases are Mijen, West Semarang, Gajah Mungkur, Candisari, South Semarang, Central Semarang, East Semarang, Gayamsari, North Semarang, Tugu. Sub-districts in group 2 with high relative risk are Ngaliyan, Gunungpati, Banyumanik, Tembalang, Pedurungan, Genuk text
institution Institut Teknologi Bandung
building Institut Teknologi Bandung Library
continent Asia
country Indonesia
Indonesia
content_provider Institut Teknologi Bandung
collection Digital ITB
language Indonesia
description Dengue fever (DHF) is an infectious disease caused by one of four different dengue viruses and is transmitted by mosquitoes, mainly Aedes aegypti and Aedes albopictus, found in tropical and subtropical regions. Dengue fever (DHF) is still one of the major public health problems in Indonesia. Along with increasing mobility and population density, the number of sufferers and the area of spread is increasing. This can be caused by climate change and low awareness to maintain environmental hygiene. This can be anticipated by modeling cases with the Autorgressive Distributed Lag (ADL) model and predicting for each region to map the risk into relative risk groups using the Localized Conditionally Autoregressive (CAR) Model. In this discussion, ADL and CAR models are used to see how far the risk of dengue cases can be formed and can anticipate the riskiest locations first. This research focuses on examining dengue cases in Semarang City from 2016 to 2021. The methodology begins with checking stationarity by looking at the P-value that is smaller than alpha, which is 0.05. The second step is to determine the appropriate model by looking at the cointegrity test and also checking the best model of the corresponding ADL. The third stage is modeling the ADL that is in accordance with the best model ADL (1,12) with ????2= 68% and making predictions for each sub-district in Semarang City. The next stage determines the value of the Moran Index to see whether autocorrelation occurs or not. Next, the best CAR modeling is used to map the risk based on the selected group. In this case, two clusters were obtained with a WAIC value of 108.43. These results indicate that the riskiest locations are divided into two, namely: Group 1 with low relative risk of DHF cases are Mijen, West Semarang, Gajah Mungkur, Candisari, South Semarang, Central Semarang, East Semarang, Gayamsari, North Semarang, Tugu. Sub-districts in group 2 with high relative risk are Ngaliyan, Gunungpati, Banyumanik, Tembalang, Pedurungan, Genuk
format Theses
author Qodri Alfairus, Muh.
spellingShingle Qodri Alfairus, Muh.
PREDICTION AND VISUALIZATION OF RELATIVE RISK MAPS USING AUTOREGRESSIVE DISTRIBUTED LAG (ADL) AND CONDITIONALLY AUTOREGRESSIVE (CAR) MODELS
author_facet Qodri Alfairus, Muh.
author_sort Qodri Alfairus, Muh.
title PREDICTION AND VISUALIZATION OF RELATIVE RISK MAPS USING AUTOREGRESSIVE DISTRIBUTED LAG (ADL) AND CONDITIONALLY AUTOREGRESSIVE (CAR) MODELS
title_short PREDICTION AND VISUALIZATION OF RELATIVE RISK MAPS USING AUTOREGRESSIVE DISTRIBUTED LAG (ADL) AND CONDITIONALLY AUTOREGRESSIVE (CAR) MODELS
title_full PREDICTION AND VISUALIZATION OF RELATIVE RISK MAPS USING AUTOREGRESSIVE DISTRIBUTED LAG (ADL) AND CONDITIONALLY AUTOREGRESSIVE (CAR) MODELS
title_fullStr PREDICTION AND VISUALIZATION OF RELATIVE RISK MAPS USING AUTOREGRESSIVE DISTRIBUTED LAG (ADL) AND CONDITIONALLY AUTOREGRESSIVE (CAR) MODELS
title_full_unstemmed PREDICTION AND VISUALIZATION OF RELATIVE RISK MAPS USING AUTOREGRESSIVE DISTRIBUTED LAG (ADL) AND CONDITIONALLY AUTOREGRESSIVE (CAR) MODELS
title_sort prediction and visualization of relative risk maps using autoregressive distributed lag (adl) and conditionally autoregressive (car) models
url https://digilib.itb.ac.id/gdl/view/77214
_version_ 1822008210645581824