To Weight or Not to Weight, That Is the Question: The Design of a Composite Indicator of Landscape Fragmentation

Composite indicators (CIs), i.e., combinations of many indicators in a unique synthetizing measure, are useful for disentangling multisector phenomena. Prominent questions concern indicators’ weighting, which implies time-consuming activities and should be properly justified. Landscape fragmentation (LF), the subdivision of habitats in smaller and more isolated patches, has been studied through the composite index of landscape fragmentation (CILF). It was originally proposed by us as an unweighted combination of three LF indicators for the study of the phenomenon in Sardinia, Italy. In this paper, we aim at presenting a weighted release of the CILF and at developing the Hamletian question of whether weighting is worthwhile or not. We focus on the sensitivity of the composite to different algorithms combining three weighting patterns (equalization, extraction by principal component analysis, and expert judgment) and three indicators aggregation rules (weighted average mean, weighted geometric mean, and weighted generalized geometric mean). The exercise provides the reader with meaningful results. Higher sensitivity values signal that the effort of weighting leads to more informative composites. Otherwise, high robustness does not mean that weighting was not worthwhile. Weighting per se can be beneficial for more acceptable and viable decisional processes.

Download Full-text

Multi-Classification Combination Algorithm Based on Logit Model and Support Vector Machine

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.734-737.2978 ◽

2013 ◽

Vol 734-737 ◽

pp. 2978-2982 ◽

Cited By ~ 1

Author(s):

Xin Lei Zhang ◽

Meng Gang Li ◽

Zuo Quan Zhang

Keyword(s):

Principal Component Analysis ◽

Support Vector Machine ◽

Regression Analysis ◽

Logit Model ◽

Prediction Accuracy ◽

Principal Component ◽

Composite Indicator ◽

Support Vector ◽

Logit Regression ◽

Multi Classification

According to the basic theories of Logit regression analysis and support vector machine, this article involves improved multi-classification combination algorithm. When applying this model, there are some innovations. First, choose optimized composite indicator as a variable through principal component analysis and get more information. Second, introduce Logit parameter model to the quadratic to increase prediction accuracy. Third, put forward a multi-classification combination model of improved Logit model with SVM to increase prediction accuracy.

Download Full-text

THE IMPACT OF REAL ESTATE INVESTMENTS ON THE PERFORMANCE OF THE ENTITIES LISTED AT THE BVB

The Annals of the University of Oradea. Economic Sciences ◽

10.47535/1991auoes30(1)019 ◽

2021 ◽

Vol 30 (30 (1)) ◽

pp. 177-186

Author(s):

Silviu Cornel Virgil Chiriac

Keyword(s):

Principal Component Analysis ◽

Real Estate ◽

Financial Performance ◽

Descriptive Analysis ◽

Composite Index ◽

Principal Component ◽

Component Analysis ◽

Multidimensional Data ◽

Data Set ◽

The Impact

The current paper is part of a wider study which aims at identifying the determining factors of the performances of the entities in the real estate field and the setting up of a composite index of the companies’ performances based on a sample of 29 companies listed at the BVB Bucharest (Bucharest Stock Exchange) in the year 2019 using one of the multidimensional data analysis techniques, the principal component analysis. The descriptive analysis, the principal component analysis for setting up the composite index of the companies performances were applied within the study in order to highlight the most important companies from the point of view of the financial performance. The descriptive analysis of the data set highlights the overview within the companies selected for analysis. The study aims at building a synthetic indicator that will show the financial performance of the companies selected based on 9 financial indicators using the principal component analysis PCA. The 9 indicators considered for the analysis were selected based on specialised articles and they are: ROA – return on assets, which reflect the company’s capacity of using its assets productively, ROE – return on equity, which measures the efficiency of use of the stockholders’ capitals, rotation of total assets, general liquidity ratio, general solvency ratio, general dent-to-equity level, net profit margin, gross return of portfolio.

Download Full-text

Assessing Landscape Fragmentation: A Composite Indicator

Sustainability ◽

10.3390/su12229632 ◽

2020 ◽

Vol 12 (22) ◽

pp. 9632

Author(s):

Andrea De Montis ◽

Vittorio Serra ◽

Amedeo Ganciu ◽

Antonio Ledda

Keyword(s):

Spatial Pattern ◽

Spatial Data ◽

Geometric Mean ◽

Algorithm Design ◽

Landscape Fragmentation ◽

Mean Value ◽

Composite Indicator ◽

Weighted Version ◽

Data Set ◽

Patch Density

The assessment and management of landscape fragmentation (LF), i.e., the subdivision of the habitat into smaller and more isolated patches, can benefit from the adoption of a composite indicator explaining, in a unique measure, the various concerns involved. However, the use of composite indicators may be affected by lack of data, subjectivity in algorithm design, and oversimplification connected to reduction to just one index. In these cases, the findings obtained might not provide the researcher with reliable information. In this paper, we design and apply the Composite Indicator of Landscape Fragmentation (CILF), a metric resuming three indicators concerning the effect on LF of transport and mobility infrastructures, human settlements, and patch density per se. The application concerns the measurement of LF spatial pattern and dynamics from 2003 to 2008 of 51 landscape units in the island of Sardinia (Italy). We considered a complete spatial data set, chose the generalized geometric mean as aggregation algorithm, and verified its robustness via sensitivity analysis of the results. We found that, in 2003 and 2008, the CILF spatial pattern shows higher values in coastal areas and has varied randomly, i.e., without a consistent tendency to converge to, or diverge from, a mean value. Overall, we demonstrate that the CILF is a powerful instrument for monitoring LF in Sardinia and advocate that it can be further implemented, following the same methodological framework, by extending the pool of indicators considered and assessing a weighted version of the composite indicator.

Download Full-text

Categorization of Chinese Dry-Cured Ham Based on Three Sticks’ Method by Multiple Sensory Technologies

Journal of Food Quality ◽

10.1155/2017/1701756 ◽

2017 ◽

Vol 2017 ◽

pp. 1-6 ◽

Cited By ~ 5

Author(s):

Dong Xia ◽

Dan Ni Zhang ◽

Shao Ting Gao ◽

Le Chen ◽

Nan Li ◽

...

Keyword(s):

Principal Component Analysis ◽

Cluster Analysis ◽

Sensory Evaluation ◽

Electronic Nose ◽

Function Analysis ◽

Principal Component ◽

Absorption Method ◽

Dry Cured Ham ◽

And Cluster Analysis ◽

Higher Sensitivity

Odor profiles of three grades of Jinhua, Xuanwei, and Rugao dry-cured hams were analyzed and distinguished by both the electronic nose and the sensory evaluation. The odor was absorbed by bamboo sticks, which is the most traditional absorption method to classify different ham grades. Then data from electronic nose was analyzed by discriminant function analysis (DFA) and cluster analysis (CA), compared with that from sensory evaluation by principal component analysis (PCA). Results showed that different grades of Jinhua, Xuanwei, and Rugao dry-cured hams could be distinguished effectively by the DFA results of electronic nose. However, sensory evaluation could not perform as well as electronic nose. It was demonstrated that intelligent sensory technology has higher sensitivity and reliability in classifying producing regions and grades of dry-cured ham.

Download Full-text

Risk of Adolescent PregnancyToward Maternal and Infant Health, Analisis of IDHS 2012

Kesmas National Public Health Journal ◽

10.21109/kesmas.v12i3.1691 ◽

2018 ◽

Vol 12 (3) ◽

Author(s):

Anni Fithriyatul Mas'udah ◽

Besral Besral ◽

Bimandra Adiputra Djaafara

Keyword(s):

Principal Component Analysis ◽

Maternal Health ◽

Adolescent Pregnancy ◽

Access To Health Care ◽

Infant Health ◽

Composite Index ◽

Principal Component ◽

Component Analysis ◽

Childbearing Age ◽

Maternal And Infant Health

AbstractImpact of adolescent pregnancy is closely related to risk of maternal and infant health. This study aimed to develop composite maternal index, infant index, and analyze impact of adolescent pregnancy on maternal and infant health. This study used Indonesian Health Demographic Survey 2012 data with a sample of 2,167 respondents qualified for inclusion (women of childbearing age giving birth to their first child since January 2007 until the survey was conducted). Analysis used Polychoric Principal Component Analysis to produce composite index of maternal and infant health, and logistic regression to determine relation of adolescent pregnancy to maternal and infant health. Results showed that index of maternal health can be determined by indicator maternal morbidity, access to health care, and health behaviors. Infant health index can be determined by infant morbidity and nutritional status indicators. Adolescents with higher education level had 5.4 times higher risk (95% CI = 1.3 – 22.0) to have worse maternal health than adult pregnancies. Adolescent pregnancy had a risk 1.3 times (95% CI = 0.9 – 1.7) to have worse infant health than adult pregnancies after controlled by occupation, socioeconomic, and region. In conclusion, the risk of adolescent pregnancy on maternal and infant health is worse than adult.AbstrakDampak kehamilan pada remaja erat kaitannya dengan risiko kesehatan ibu dan bayi. Penelitian ini bertujuan untuk mengembangkan komposit indeks kesehatan ibu, indeks kesehatan bayi dan untuk menelaah pengaruh kehamilan remaja terhadap kesehatan ibu dan bayi. Penelitian ini menggunakan data Survei Demografi Kesehatan Indonesia tahun 2012, dengan sampel 2.167 responden yang memenuhi syarat inklusi (wanita usia subur yang pernah melahirkan anak pertama sejak Januari 2007 sampai survei dilaksanakan). Analisis menggunakan Polychoric Principal Component Analysis untuk menentukan komposit indeks kesehatan ibu dan bayi, serta regresi logistik untuk mengetahui besar hubungan kehamilan remaja terhadap kesehatan ibu dan bayi. Kesehatan ibu dapat ditentukan melalui indikator status kesakitan ibu, akses layanan kesehatan, dan perilaku kesehatan. Kesehatan bayi dapat ditentukan melalui indikator status kesakitan bayi dan status gizi bayi. Kehamilan remaja dengan status pendidikan tinggi memiliki risiko kesehatan ibu lebih buruk 5,4 kali (95% CI = 1,3– 22,0) dibandingkan pada kehamilan dewasa. Kehamilan remaja memiliki risiko kesehatan bayi lebih buruk 1,3 kali (95% CI = 0,9 – 1,7) dibandingkan kehamilan dewasa setelah dikontrol oleh pekerjaan, sosial ekonomi, dan wilayah. Penelitian ini menyimpulkan bahwa risiko kehamilan remaja pada kesehatan ibu dan bayi lebih buruk dibandingkan usia dewasa.

Download Full-text

Indonesian Provinces SDGs Composite Index: Lampung Province Analysis

Economics and Finance in Indonesia ◽

10.47291/efi.v67i1.819 ◽

2021 ◽

Vol 67 (1) ◽

pp. 1

Author(s):

Reny Andriati ◽

Arief Anshory Yusuf

Keyword(s):

Principal Component Analysis ◽

Sustainable Development Goals ◽

National Level ◽

Composite Index ◽

Principal Component ◽

Factor Loading ◽

Component Analysis ◽

Provincial Level ◽

Development Goals ◽

Economic Survey

Publications of Sustainable Development Goals (SDGs) have mainly been conducted at a national level and separately for each goal. No prior research has been done on SDGs composite index at a provincial level in Indonesia. It is necessary to create a composite index that presents a single value at the provincial level to enable regional evaluation. The Indonesia Province SDGs composite index is developed from indicators based on Statistics Indonesia gathered from several publications. The data sources are the National Socio-Economic Survey (Susenas) and the Basic Health Research (Riskesdas) which were linked surveys held in 2018. Principal Component Analysis and Factor Analysis are used as the methods to select the indicators of the SDGs. Those selected indicators are then normalized using the min-max method and subsequently weighted using factor loading derived from the principal component analysis. Finally, the indicators are aggregated using an arithmetic mean to determine the composite index. The Indonesia Province SDGs composite index is an approach to measure achievement of SDGs agenda. In addition, each goal achievement is summarized as a goal index. The SDGs composite index for Lampung Province is 52.2%, meaning that Lampung Province is 52.2% of the way to fully achieving the SDGs, according to the measures used to calculate this index. The findings on goal index suggest that development is highly requested on public services such as housing and water supply.

Download Full-text

Consciousness & Brain Functional Complexity in Propofol Anaesthesia

10.1101/680447 ◽

2019 ◽

Cited By ~ 1

Author(s):

TF Varley ◽

A Luppi ◽

I Pappas ◽

L Naci ◽

R Adapa ◽

...

Keyword(s):

Complex System ◽

Principal Component Analysis ◽

Anaesthetic Agent ◽

Principal Component ◽

Topological Dimension ◽

Functional Complexity ◽

Temporal Measures ◽

Process Complexity ◽

The Brain ◽

Higher Sensitivity

AbstractThe brain is possibly the most complex system known to mankind, and its complexity has been called upon to explain the emergence of consciousness. However, complexity can take many forms: here, we investigate measures of algorithmic and process complexity in both the temporal and topological dimension, testing them on functional MRI data obtained from individuals undergoing various levels of sedation with the anaesthetic agent propofol, in two separate datasets. We demonstrate that the various measures are differently able to discriminate between levels of sedation, with temporal measures showing higher sensitivity. Further, we show that all measures are strongly related to a single underlying construct explaining most of the variance, as assessed by Principal Component Analysis, which we interpret as a measure of overall complexity of our data. This overall complexity was also able to discriminate between levels of sedation, supporting the hypothesis that consciousness is related to complexity - independent of how the latter is measured.

Download Full-text

Analog Circuits Test by Using Principal Component Analysis

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.278-280.709 ◽

2013 ◽

Vol 278-280 ◽

pp. 709-713

Author(s):

Chao Jie Zhang ◽

Guang Hui Chang

Keyword(s):

Principal Component Analysis ◽

Analog Circuits ◽

Principal Component ◽

Component Analysis ◽

Test Methods ◽

Component Model ◽

Suitable Candidate ◽

Principal Component Model ◽

Other Test Methods ◽

Higher Sensitivity

In view of the difficulties caused by determining threshold for analog circuits test, a method based on principal component analysis (PCA) of node voltages was proposed to overcome these difficulties. At first, the principal component model of fault-free circuits was constructed. Then the circuits-under-test was compared with the principal component model to calculate the statistic for fault detection. The proposed method was used to test the output signal amplifying circuit, which is used in the ultrasonic liquid sensor. The testing results show that the PCA based method has a higher sensitivity than other test methods. And the proposed method can overcome the difficulties in determining threshold by the expert’s empirical knowledge. These make it a suitable candidate for analog circuits test.

Download Full-text

Measuring Economic Integration in the Asia-Pacific Region: A Principal Components Approach

Asian Economic Papers ◽

10.1162/asep_a_00009 ◽

2010 ◽

Vol 9 (2) ◽

pp. 121-143 ◽

Cited By ~ 12

Author(s):

Bo Chen ◽

Yuen Pau Woo

Keyword(s):

Principal Component Analysis ◽

Economic Integration ◽

Composite Index ◽

Principal Component ◽

Asia Pacific ◽

Pacific Region ◽

Economic Convergence ◽

Asia Pacific Region ◽

Second Stage ◽

Macroeconomic Indicators

This paper measures economic integration in the Asia-Pacific (AP) region using a composite index. The weights of the index are obtained from a two-stage principal component analysis. In the first stage, we obtain a convergence index to measure the extent of convergence among the main macroeconomic indicators of a sample of AP economies. In the second stage, we use indicators of trade, FDI, and tourism, as well as the convergence index, to compute the weights for the composite index. We found that economic convergence in the AP region increased until 1998 but has since fallen back. The integration of trade, investment, and people flows increased between 1990 and 2000, weakened slightly to 2003, and has since picked up again. Among the 17 sample economies, Singapore, Hong Kong, and Chinese Taipei are the most integrated with the AP region and Indonesia and China are the least integrated.

Download Full-text