scholarly journals Using Artificial Neural networks and Optimal Scaling Model to Forecast Agriculture Commodity Price: An Ecological-economic Approach

2021 ◽  
pp. 29-55
Author(s):  
Roberto Louis Forestal ◽  
Shih-Ming Pi
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Food System ◽  
Commodity Price ◽  
Optimal Scaling ◽  
Economic Approach ◽  
Pricing Model ◽  
Input Output ◽  
Ecological Economic ◽  
Artificial Neural

This research paper employs input-output pricing model based on ecological-economic approach to investigate the impacts of internal factors as well as external forces on agriculture commodities. To empirically test our model, we select two different methodologies such as the optimal scaling regression with nonlinear transformations and feedforward artificial neural networks. Our sample includes data related to price of agriculture and energy commodities (cocoa, coffee and crude oil), production of crops and livestock, emissions of greenhouse gases (GHG) from agriculture from 1961 to 2019. Results find a bidirectional relationship between cocoa price and coffee price explaining by the fact that commodity-dependent countries often use kindred production landscapes and similar supply chain management when dealing with coffee and cocoa. Therefore, effect of supply side shocks may be transmitted from one market to another. We also present evidence that greenhouse gas emissions have strong effect on commodity price, thus we encourage an integrated approach including both concrete technological and proactive managerial measures in order to mitigate global warming impacts on the food system. We believe that these findings will be of interest to commodity producers, asset managers and academics who look a better understanding of the dynamics of commodity markets. JEL classification numbers: C50, Q02, Q57. Keywords: Agriculture commodity, Input-output pricing model, Ecological-economic approach, Artificial neural networks, Optimal scaling regression.

scholarly journals IDENTIFIKASI POLA KARAKTER ANAK DENGAN ALGORITMA PERCEPTRON

2019 ◽  
Vol 3 (1) ◽  
pp. 86
Author(s):  
Sri Rahmadhany
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
Human Brain ◽  
Computational Method ◽  
Input Output ◽  
Perceptron Algorithm ◽  
Behavioral Variables ◽  
Artificial Neural

Abstract - Artificial Neural Network is a computational method that works like a human brain. The Perceptron algorithm is one method that exists in Artificial Neural Networks. The research carried out was the identification of children's character patterns using the Perceptron algorithm. The Perceptron algorithm is very reliable in recognizing patterns, one of which is the child's character pattern as was done in this study. The Perceptron algorithm identifies the character patterns of children through three inputs and two outputs. The three outputs are taken from nature variables, attitude variables and behavioral variables. The output is four human temperaments according to Hipocrates, namely sanguin, melancholy, choleric and plegamatic. All inputs and outputs will be converted into binary numbers to be trained with Matlab software.Keywords - Artificial Neural Networks, Perceptron Algorithms, child character patterns, input, output, binary numbers. Abstrak - Jaringan Syaraf Tiruan merupakan salah satu metode komputasi yang dapat bekerja seperti layaknya otak manusia. Algortima Perceptron merupakan salah satu metode yang ada pada Jaringan Syaraf Tiruan. Penelitian yang dilakukan adalah identifikasi pola karakter anak dengan menggunakan algoritma Perceptron. Algoritma Perceptron sangat handal dalam mengenali pola salah satunya yaitu pola karakter anak seperti yang dilakukan dalam penelitian ini. Algoritma Perceptron mengidentifikasi pola karakter anak melalui tiga input dan dua output. Tiga output tersebut diambil dari variabel sifat, variabel sikap dan variabel tingkah laku. Adapun output merupakan empat temperamen manusia menurut Hipocrates yaitu sanguin, melankolis, koleris dan plegamatis. Seluruh input dan output akan diubah menjadi bilangan biner untuk dilatih dengan software Matlab.Kata Kunci - Jaringan Syaraf Tiruan, Algoritma Perceptron, pola karakter anak, input, output, bilangan biner.

scholarly journals Estimation of Residential Property Market Price: Comparison of Artificial Neural Networks and Hedonic Pricing Model

2020 ◽  
Vol 4 (6) ◽  
pp. 530-538
Author(s):  
Michaela Štubňová ◽  
Marta Urbaníková ◽  
Jarmila Hudáková ◽  
Viera Papcunová
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Real Estate ◽  
Learning Algorithm ◽  
Market Price ◽  
Residential Property ◽  
Pricing Model ◽  
Property Market ◽  
Price Estimation ◽  
Artificial Neural

The correct real estate property price estimation is significant not only in the real estate market but also in the banking sector for collateral loans and the insurance sector for property insurance. The paper focuses on both traditional and advanced methods for real estate property valuation. Attention is paid to the analysis of the accuracy of valuation models. From traditional methods, a regression model is used for residential property price estimation, which represents the hedonic approach. Modern advanced valuation methods are represented by the artificial neural network, which is one of the soft computing techniques. The results of both methods in residential property market price estimation are compared. The analysis is performed using data on residential properties sold on the real estate market in the city of Nitra in the Slovak Republic. To estimate the residential property prices, artificial neural networks trained with the Levenberg-Marquart learning algorithm, the Bayesian Regularization learning algorithm, and the Scaled Conjugate Gradient learning algorithm, and the regression pricing model are used. Among the constructed neural networks, the best results are achieved with networks trained with the Regularization learning algorithm with two hidden layers. Its performance is compared with the performance of the regression pricing model, and it can state that artificial neural networks can considerably improve prediction accuracy in the estimation of residential property market price. Doi: 10.28991/esj-2020-01250 Full Text: PDF

Learning in Feed-Forward Artificial Neural Networks II

2011 ◽  
pp. 1012-1017
Author(s):  
Lluís A. Belanche Muñoz
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Input Output ◽  
Rbf Networks ◽  
Learning Rules ◽  
The Neural Network ◽  
Artificial Neural ◽  
Considerable Work ◽  
Derivative Information

Supervised Artificial Neural Networks (ANN) are information processing systems that adapt their functionality as a result of exposure to input-output examples. To this end, there exist generic procedures and techniques, known as learning rules. The most widely used in the neural network context rely in derivative information, and are typically associated with the Multilayer Perceptron (MLP). Other kinds of supervised ANN have developed their own techniques. Such is the case of Radial Basis Function (RBF) networks (Poggio & Girosi, 1989). There has been also considerable work on the development of adhoc learning methods based on evolutionary algorithms.

Sign in / Sign up

Export Citation Format

Share Document