Pemanfaatan CFSRv2 untuk Statistical Downscaling menggunakan Principal Component Regression dan Partial Least Square

2019 ◽  
Vol 8 (1) ◽  
Author(s):  
Khairunnisa Khairunnisa ◽  
Rizka Pitri ◽  
Victor P Butar-Butar ◽  
Agus M Soleh
Keyword(s):  
Statistical Downscaling ◽  
General Circulation ◽  
Principal Component Regression ◽  
Circulation Model ◽  
Principal Component ◽  
Meteorological Station ◽  
Grid Size ◽  
Partial Least Square ◽  
Least Square ◽  
Output Data

This research used CFSRv2 data as output data general circulation model. CFSRv2 involves some variables data with high correlation, so in this research is using principal component regression (PCR) and partial least square (PLS) to solve the multicollinearity occurring in CFSRv2 data. This research aims to determine the best model between PCR and PLS to estimate rainfall at Bandung geophysical station, Bogor climatology station, Citeko meteorological station, and Jatiwangi meteorological station by comparing RMSEP value and correlation value. Size used was 3×3, 4×4, 5×5, 6×6, 7×7, 8×8, 9×9, and 11×11 that was located between (-40) N - (-90) S and 1050 E -1100 E with a grid size of 0.5×0.5 The PLS model was the best model used in stastistical downscaling in this research than PCR model because of the PLS model obtained the lower RMSEP value and the higher correlation value. The best domain and RMSEP value for Bandung geophysical station, Bogor climatology station, Citeko meteorological station, and Jatiwangi meteorological station is 9 × 9 with 100.06, 6 × 6 with 194.3, 8 × 8 with 117.6, and 6 × 6 with 108.2, respectively.

scholarly journals Application of FTIR Spectroscopy and HPLC Combined with Multivariate Calibration for Analysis of Xanthones in Mangosteen Extracts

2020 ◽  
Vol 88 (3) ◽  
pp. 35
Author(s):  
Endjang Prebawa Tejamukti ◽  
Widiastuti Setyaningsih ◽  
Irnawati ◽  
Budiman Yasir ◽  
Gemini Alam ◽  
...  
Keyword(s):  
Quantitative Analysis ◽  
Multivariate Calibration ◽  
Principal Component Regression ◽  
Principal Component ◽  
Ethanolic Extract ◽  
Ftir Spectra ◽  
Partial Least Square ◽  
Least Square ◽  
Coefficient Of Determination ◽  
Garcinia Mangostana

Mangosteen, or Garcinia mangostana L., has merged as an emerging fruit to be investigated due to its active compounds, especially xanthone derivatives such as α -mangostin (AM), γ-mangostin (GM), and gartanin (GT). These compounds had been reported to exert some pharmacological activities, such as antioxidant and anti-inflammatory, therefore, the development of an analytical method capable of quantifying these compounds should be investigated. The aim of this study was to determine the correlation between FTIR spectra and HPLC chromatogram, combined with chemometrics for quantitative analysis of ethanolic extract of mangosteen. The ethanolic extract of mangosteen pericarp was prepared using the maceration technique, and the obtained extract was subjected to measurement using instruments of FTIR spectrophotometer at wavenumbers of 4000–650 cm−1 and HPLC, using a PDA detector at 281 nm. The data acquired were subjected to chemometrics analysis of partial least square (PLS) and principal component regression (PCR). The result showed that the wavenumber regions of 3700–2700 cm−1 offered a reliable method for quantitative analysis of GM with coefficient of determination (R2) 0.9573 in calibration and 0.8134 in validation models, along with RMSEC value of 0.0487% and RMSEP value 0.120%. FTIR spectra using the second derivatives at wavenumber 3700–663 cm−1 with coefficient of determination (R2) >0.99 in calibration and validation models, along with the lowest RMSEC value and RMSEP value, were used for quantitative analysis of GT and AM, respectively. It can be concluded that FTIR spectra combined with multivariate are accurate and precise for the analysis of xanthones.

scholarly journals COMPARATIVE STUDY OF FATTY ACID PROFILES IN PATIN (PANGASIUS MICRONEMUS) AND GABUS (CHANNA STRIATA) FISH OIL AND ITS AUTHENTICATION USING FTIR SPECTROSCOPY COMBINED WITH CHEMOMETRICS

2019 ◽  
pp. 55-60
Author(s):  
ANGGITA ROSIANA PUTRI ◽  
ABDUL ROHMAN ◽  
SUGENG RIYANTO
Keyword(s):  
Fatty Acids ◽  
Fatty Acid ◽  
Fish Oil ◽  
Ftir Spectroscopy ◽  
Principal Component Regression ◽  
Principal Component ◽  
Partial Least Square ◽  
Least Square ◽  
Coefficient Of Determination ◽  
Channa Striata

Objective: The aims of this research were to analyse the fatty acids contained in Patin (Pangasius micronemus) and Gabus (Channa striata) fish oils also its authentication using FTIR spectroscopy combined with chemometrics. Methods: Patin fish oil (PFO) was extracted from patin flesh using the maceration method with petroleum benzene as the solvent, while gabus fish oil (GFO) was purchased from the market in Yogyakarta. The analysis of fatty acid was done using gas chromatography–flame ionization detector (GC-FID). The authentication was performed using FTIR spectrophotometer and chemometrics methods. Principal component analysis (PCA) was used to determine the proximity of oils based on the characteristic similarity. The quantification of adulterated PFO was performed using multivariate calibrations, partial least square (PLS) and principal component regression (PCR). The classification between authentic oils and those adulterated used discriminant analysis (DA). Results: The level of saturated and polyunsaturated fatty acids in PFO is higher than in GFO. The PLS and PCR methods using the second derivative spectra at wavenumbers of 666–3050 cm-1 offered the highest values of coefficient of determination (R2) and lowest root means the square error of calibration (RMSEC) and root mean square error of prediction (RMSEP). Conclusion: The PCA method was successfully used to determine the proximity of oils. Among oils studied, PFO has a similarity fatty acid composition with GFO. The DA method was able to screen pure PFO from adulterated PFO without any misclassification reported. FTIR spectroscopy in combined with chemometrics can be used for authentication and quantification.

scholarly journals PENERAPAN KEMOMETRIK PADA METODE CITRA DIGITAL UNTUK ANALISIS ION MERKURI (II) DENGAN INDIKATOR NANOPARTIKEL PERAK

Alotrop ◽  
2019 ◽  
Vol 3 (1) ◽  
Author(s):  
Angga Aprian Dinata ◽  
M. Lutfi Firdaus ◽  
Rina Elvia
Keyword(s):  
Quantitative Analysis ◽  
Digital Image ◽  
Limit Of Detection ◽  
Principal Component Regression ◽  
Image Data ◽  
Principal Component ◽  
Partial Least Square ◽  
Least Square ◽  
Image Method ◽  
Rgb Values

Digital image method in quantitative analysis usually uses one of the RGB primary color components (Red, Green, Blue), so that not all digital image data can be extracted. Then needed a method that can render the whole RGB values as variables in quantitative analysis are known as chemometric. This research aims to know the influence of the application of chemometric against the sensitivity of the digital image. Chemometry method used is the Principal Component Regression (PCR) and Partial Least Square (PLS) using Unscramber X software from Camo software, USA.. This method is applied for the quantitative analysis of Mercury (II) ion with silver nanoparticles (NPP) immobilization on filter paper indicator. The research results showed that chemometric has a good influence against the level of the Limit of Detection (LOD) of the digital image, where the level of LOD with chemometric application of the Principal Component Regression (PCR) is 0.4311 ppb, and Partial Least Square (PLS) is  0.4310 ppb smaller than without the application of chemometric Single Linear Regression (SLR) at 0.837 ppb. 

scholarly journals Determination of acid, peroxide, and saponification value in patin fish oil by FTIR spectroscopy combined with chemometrics

Food Research ◽  
2020 ◽  
Vol 4 (5) ◽  
pp. 1758-1766
Author(s):  
A.R. Putri ◽  
A. Rohman ◽  
W. Setyaningsih ◽  
S. Riyanto
Keyword(s):  
Fish Oil ◽  
Ftir Spectroscopy ◽  
Principal Component Regression ◽  
Principal Component ◽  
Acid Value ◽  
Partial Least Square ◽  
Least Square ◽  
Coefficient Of Determination ◽  
Saponification Value

Simple, rapid, and reproducible methods for determining the acid value (AV), peroxide value (PV), and saponification value (SV) of patin fish oil (PFO) were developed using Fourier Transform Infrared (FTIR) spectroscopy combined with chemometrics of Principal Component Regression (PCR) and Partial Least Square (PLS). The relationship between actual values was determined using AOCS method and predicted value was determined with FTIR spectroscopy and chemometrics. From the validation work, the high coefficient of determination (R2 ) reached up to > 0.99. This study concluded that by means of FTIR spectra that combined with PCR and PLS technique can be used to determine AV, PV, and SV of PFO.

scholarly journals INFRARED SPECTROSCOPY AND MULTIVARIATE CALIBRATION FOR THE RAPID QUANTIFICATION OF FREE FATTY ACID CONTENT IN PANGASIUS HYPOPTHALMUS OIL

2018 ◽  
Vol 10 (6) ◽  
pp. 199
Author(s):  
Lisa Andina ◽  
Revita Saputri ◽  
Aristha Novyra Putri ◽  
Abdul Rohman
Keyword(s):  
Fatty Acid ◽  
Ftir Spectroscopy ◽  
Multivariate Calibration ◽  
Fatty Acid Content ◽  
Principal Component Regression ◽  
Principal Component ◽  
Partial Least Square ◽  
Least Square ◽  
Coefficient Of Determination ◽  
Stretching Vibration

Objective: The objective of this study was to evaluate the capability of fourier transform infrared (FTIR) spectroscopy in combination with multivariate calibration for prediction of free fatty acids (FFA) in Pangasius hypopthalmus (P. hypopthalmus) oil.Methods: FFA content in P. hypopthalmus oil was determined by attenuated total reflectance-FTIR spectroscopy. P. hypopthalmus oil derived from Pangasius’s meat (MP), and Pangasius’s liver and fat (LFP) were subjected to heat treatments. Determination of FFA content in P. hypopthalmus oil’s was performed by gas chromatography-flame ionization detector.Results: Oleic acid was found to be the main fatty acid component in P. hypopthalmus oil. FTIR spectra of P. hypopthalmus oil has 3 main peaks, C-H bonds of cis-form of fatty acid showed the stretching vibration, symmetric and asymmetric vibrations of the C-H2 and C-H3 aliphatic group and vibrations of the carbonyl (C=O) ester derived from the oil triacylglycerols. Principal component regression (PCR) model showed a better performance than the partial least square (PLS) model. PCR at wavenumbers of 1200-1000 cm-1 with first derivative treatment was chosen for FFA prediction, which resulted in a coefficient of determination (R2) value of 0.9417, root means square error of calibration (RMSEC) of 0.725%, and root mean square error of prediction (RMSEP) value of 2.40%, respectively.Conclusion: FTIR spectroscopy combined with PCR can be used as an alternative method for analysis of fatty acid contents.

scholarly journals MODEL OUTPUT STATISTICS DENGAN PRINCIPAL COMPONENT REGRESSION, PARTIAL LEAST SQUARE REGRESSION, DAN RIDGE REGRESSION UNTUK KALIBRASI PRAKIRAAN CUACA JANGKA PENDEK

2021 ◽  
Vol 10 (3) ◽  
pp. 355
Author(s):  
NISWATUL QONA’AH ◽  
HASIH PRATIWI ◽  
YULIANA SUSANTI
Keyword(s):  
Ridge Regression ◽  
Principal Component Regression ◽  
Principal Component ◽  
Partial Least Square ◽  
Least Square ◽  
Partial Least Square Regression ◽  
Model Output ◽  
Mean Square ◽  
Least Square Regression ◽  
Model Output Statistics

Penelitian ini merupakan upaya pengembangan Model Output Statistics (MOS) yang akan digunakan sebagai alat kalibrasi prakiraan cuaca jangka pendek. Informasi mengenai prakiraan cuaca yang akurat diharapkan dapat meminimalkan risiko kecelakaan yang disebabkan oleh cuaca, khususnya dalam bidang transportasi udara dan laut. Metode yang akan dikembangkan mencakup beberapa stasiun pengamatan cuaca di Indonesia. MOS merupakan sebuah metode berbasis regresi yang mengoptimalkan hubungan antara observasi cuaca dan luaran model Numerical Weather Predictor (NWP). Beberapa masalah yang muncul kaitannya dengan MOS adalah; mereduksi dimensi luaran NWP, mendapatkan variabel prediktor yang mampu menjelaskan variabilitas variabel respon, dan menentukan metode statistik yang sesuai dengan karakteristik data, sehingga dapat menggambarkan hubungan antara variabel respon dan variabel prediktor. Tujuan dari penelitian ini yaitu untuk mendapatkan pemodelan MOS yang sesuai untuk variabel respon suhu maksimum, suhu minimum, dan kelembapan udara. Metode regresi yang digunakan adalah Principal Component Regression (PCR), Partial Least Square Regression (PLSR), dan ridge regression. Selanjutnya, model MOS yang terbentuk divalidasi dengan kriteria Root Mean Square Error (RMSE) dan Percentage Improval (IM%). MOS mampu mengoreksi bias prakiraan NWP hingga lebih dari 50%. Berdasarkan RMSE terkecil pada penelitian ini, suhu maksimum lebih akurat diprakirakan menggunakan model PLSR, sementara suhu minimum dan kelembapan udara lebih akurat diprakirakan menggunakan ridge regression.Kata Kunci: cuaca, MOS, NWP.

scholarly journals Development of Advanced Chemometric-Assisted Spectrophotometric Methods for the Determination of Cromolyn Sodium and Its Alkaline Degradation Products

Molecules ◽  
2020 ◽  
Vol 25 (24) ◽  
pp. 5953
Author(s):  
Noha M. El Zahar ◽  
Mariam M. Tadros ◽  
Bassam M. Ayoub
Keyword(s):  
Degradation Products ◽  
Principal Component Regression ◽  
Principal Component ◽  
Pharmaceutical Products ◽  
Partial Least Square ◽  
Cromolyn Sodium ◽  
Least Square ◽  
Alkaline Degradation ◽  
Stability Indicating

Advanced and sensitive spectrophotometric and chemometric analytical methods were successfully established for the stability-indicating assay of cromolyn sodium (CS) and its alkaline degradation products (Deg1 and Deg2). Spectrophotometric mean centering ratio spectra method (MCR) and chemometric methods, including principal component regression (PCR) and partial least square (PLS-2) methods, were applied. Peak amplitudes after MCR at 367.8 nm, 373.8 nm and 310.6 nm were used within linear concentration ranges of 2–40 µg mL−1, 5–40 µg mL−1 and 10–100 µg mL−1 for CS, Deg1 and Deg2, respectively. For PCR and PLS-2 models, a calibration set of eighteen mixtures and a validation set of seven mixtures were built for the simultaneous determination of CS, Deg1 and Deg2 in the ranges of 5–13 µg mL−1, 8–16 µg mL−1, and 10–30 µg mL−1, respectively. The authors emphasize the importance of a stability-indicating strategy for the investigation of pharmaceutical products.

scholarly journals PEMODELAN STATISTICAL DOWNSCALING DENGAN PEUBAH DUMMY BERDASARKAN TEKNIK CLUSTER HIERARKI DAN NON- HIERARKI UNTUK PENDUGAAN CURAH HUJAN

2019 ◽  
Vol 3 (3) ◽  
pp. 295-309
Author(s):  
Sitti Sahriman ◽  
Anisa Kalondeng ◽  
Vieri Koerniawan
Keyword(s):  
Statistical Downscaling ◽  
Principal Component Regression ◽  
Circulation Model ◽  
Principal Component ◽  
Hierarchical Cluster ◽  
Statistical Technique ◽  
Rainfall Data ◽  
Global Circulation Model ◽  
Cluster Technique ◽  
Dummy Variables

Statistical downscaling (SD) is a statistical technique used to predict local scale rainfall based on global atmospheric circulation. The global scale climate variable used is precipitation from GCM (Global Circulation Model). However, the precipitation data of GCM outputs have a large dimension, giving rise to multicollinearity in the data. This problem is handled by the Principal Component Regression (PCR) method. In addition, the SD models have heterogeneous error variances. The dummy variable is added to the PCR models to solve the problem. Hierarchical (k-means) and non-hierarchical cluster techniques (average linkage, median linkage, and ward linkage) are used in modeling to determine rainfall data groups. Furthermore, the group formed is the basis of the formation of dummy variables. This study aims to estimate local rainfall data in Pangkep district as a salt-producing area in South Sulawesi. There are 4 dummy variables based on the 5 groups formed. Dummy variables are able to improve predictions from the PCR models. R2 values of the PCR-dummy models (ranging from 89.89% to 95.58%) are relatively higher than the PCR models (ranging from 55.87% to 57.61%). This result is also consistent with the model validation stage. The PCR-dummy models based on non-hierarchical cluster techniques (k-means) are better than the PCR-dummy models based on cluster hierarchy techniques. In general, the best model is the PCR-dummy model of the non-hierarchical cluster technique (k-means ) and involves 4 main components.

Sign in / Sign up

Export Citation Format

Share Document