Search results for: panel data regression models

Authors: M. A. C. S. Sampath Fernando, James M. Curran, Renate Meyer

Abstract:

Forensic DNA analysis has received much attention over the last three decades, due to its incredible usefulness in human identification. The statistical interpretation of DNA evidence is recognised as one of the most mature fields in forensic science. Peak heights in an Electropherogram (EPG) are approximately proportional to the amount of template DNA in the original sample being tested. A stutter is a minor peak in an EPG, which is not masking as an allele of a potential contributor, and considered as an artefact that is presumed to be arisen due to miscopying or slippage during the PCR. Stutter peaks are mostly analysed in terms of stutter ratio that is calculated relative to the corresponding parent allele height. Analysis of mixture profiles has always been problematic in evidence interpretation, especially with the presence of PCR artefacts like stutters. Unlike binary and semi-continuous models; continuous models assign a probability (as a continuous weight) for each possible genotype combination, and significantly enhances the use of continuous peak height information resulting in more efficient reliable interpretations. Therefore, the presence of a sound methodology to distinguish between stutters and real alleles is essential for the accuracy of the interpretation. Sensibly, any such method has to be able to focus on modelling stutter peaks. Bayesian nonparametric methods provide increased flexibility in applied statistical modelling. Mixture models are frequently employed as fundamental data analysis tools in clustering and classification of data and assume unidentified heterogeneous sources for data. In model-based clustering, each unknown source is reflected by a cluster, and the clusters are modelled using parametric models. Specifying the number of components in finite mixture models, however, is practically difficult even though the calculations are relatively simple. Infinite mixture models, in contrast, do not require the user to specify the number of components. Instead, a Dirichlet process, which is an infinite-dimensional generalization of the Dirichlet distribution, is used to deal with the problem of a number of components. Chinese restaurant process (CRP), Stick-breaking process and Pólya urn scheme are frequently used as Dirichlet priors in Bayesian mixture models. In this study, we illustrate an infinite mixture of simple linear regression models for modelling stutter ratio and introduce some modifications to overcome weaknesses associated with CRP.

Keywords: Chinese restaurant process, Dirichlet prior, infinite mixture model, PCR stutter

Procedia PDF Downloads 306

Authors: Amitha Puranik, V. S. Binu, Seena Biju

Abstract:

Missing data is a common problem in spatial analysis especially at the aggregate level. Missing can either occur in covariate or in response variable or in both in a given location. Many missing data techniques are available to estimate the missing data values but not all of these methods can be applied on spatial data since the data are autocorrelated. Hence there is a need to develop a method that estimates the missing values in both response variable and covariates in spatial data by taking account of the spatial autocorrelation. The present study aims to develop a model to estimate the missing data points at the aggregate level in spatial data by accounting for (a) Spatial autocorrelation of the response variable (b) Spatial autocorrelation of covariates and (c) Correlation between covariates and the response variable. Estimating the missing values of spatial data requires a model that explicitly account for the spatial autocorrelation. The proposed model not only accounts for spatial autocorrelation but also utilizes the correlation that exists between covariates, within covariates and between a response variable and covariates. The precise estimation of the missing data points in spatial data will result in an increased precision of the estimated effects of independent variables on the response variable in spatial regression analysis.

Keywords: spatial regression, missing data estimation, spatial autocorrelation, simulation analysis

Procedia PDF Downloads 347

Authors: Zahid Ahmad, Nauman Ali

Abstract:

This study has estimated the trade between USA and individual members of European Union using Gravity Model of Trade as The USA has a complex trade relationship with the European countries consist of a large number of consumers, which make USA dependent on EU for major of its total world trade. However, among the member of EU, the trade potential of USA with individual members of EU is not known. Panel data techniques e.g. Random Effect, Fixed Effect and Pooled Panel have been applied to secondary quantitative data to analyze the Trade between USA and EU. Trade Potential of USA with individual members of EU has been obtained using the ratio of Actual trade of USA with EU members and the trade as predicted by Gravity Model. The Study concluded that the USA has greater trade potential with 16 members of EU, including Croatia, Portugal and United Kingdom on top. On the other hand, Finland, Ireland, and France are the top countries with which the USA has exhaustive trade potential.

Keywords: analytical technique, economic, gravity, international trade, significant

Procedia PDF Downloads 272

Authors: Suzan Anwar, Utkarsh Goyal

Abstract:

Heart rate (HR) is a predictor of cardiovascular, cerebrovascular, and all-cause mortality in the general population, as well as in patients with cardio and cerebrovascular diseases. Machine learning (ML) significantly improves the accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment while avoiding unnecessary treatment of others. This research examines relationship between the individual's various heart health inputs like age, sex, cp, trestbps, thalach, oldpeaketc, and the likelihood of developing heart disease. Machine learning techniques like logistic regression and decision tree, and Python are used. The results of testing and evaluating the model using the Heart Failure Prediction Dataset show the chance of a person having a heart disease with variable accuracy. Logistic regression has yielded an accuracy of 80.48% without data handling. With data handling (normalization, standardscaler), the logistic regression resulted in improved accuracy of 87.80%, decision tree 100%, random forest 100%, and SVM 100%.

Keywords: heart rate, machine learning, SVM, decision tree, logistic regression, random forest

Procedia PDF Downloads 117

Authors: Mouldi Djelassi, Jamel Boukhatem

Abstract:

Background: A relation has been established between Islamic banks' activities and interest rates. The aim of this study was to explore the impact of interest rates on the deposits and loans held by Islamic and conventional banks in Saudi Arabia. Methods: A time series data was performed over the period 2008Q1-2020Q2 on eight conventional banks and four Islamic banks. The impacts of interest rate shocks on deposits and loans were identified through panel vector autoregressive models. Results: Impulse response function analysis showed that increasing interest rates reduce loans and conventional deposits. For Islamic banks, deposits are more affected by interest rates than lending. Variance decomposition analysis revealed that deposits contribute to 61% of the Islamic financing variation and only 25% of the conventional loans. Conclusion: Interest rates impacted Islamic banks especially through deposits, which is inconsistent with the theoretical framework. Islamic deposits played an important role in Islamic financing variation and may provide to be a channel for the transmission of the monetary policy in a dual banking system. Monetary policy in Saudi Arabia works in part through “credits” (conventional bank credits) as well as through “money” (conventional and Islamic bank deposits).

Keywords: Islamic banking, interest rates, monetary policy transmission, panel VAR

Procedia PDF Downloads 78

Authors: Darren Zou

Abstract:

Forecasting the stock market is a very challenging task. Various economic indicators such as GDP, exchange rates, interest rates, and unemployment have a substantial impact on the stock market. Time series models are the traditional methods used to predict stock market changes. In this paper, a machine learning method, Bayesian Additive Regression Tree (BART) is used in predicting stock market indexes based on multiple economic indicators. BART can be used to model heterogeneous treatment effects, and thereby works well when models are misspecified. It also has the capability to handle non-linear main effects and multi-way interactions without much input from financial analysts. In this research, BART is proposed to provide a reliable prediction on day-to-day stock market activities. By comparing the analysis results from BART and with time series method, BART can perform well and has better prediction capability than the traditional methods.

Keywords: BART, Bayesian, predict, stock

Procedia PDF Downloads 99

Authors: Hamidreza Ghazvinian, Khosro Ghazvinian, Touba Khodaiean

Abstract:

The importance of evaporation estimation in water resources and agricultural studies is undeniable. Pan evaporation are used as an indicator to determine the evaporation of lakes and reservoirs around the world due to the ease of interpreting its data. In this research, intelligent models were investigated in estimating pan evaporation on a daily basis. Shahroud and Mayamey were considered as the studied cities. These two cities are located in Semnan province in Iran. The mentioned cities have dry weather conditions that are susceptible to high evaporation potential. Meteorological data of 11 years of synoptic stations of Shahrood and Mayamey cities were used. The intelligent models used in this study are Artificial Neural Network (ANN), Least Squares Support Vector Machine (LSSVM), and M5 tree models. Meteorological parameters of minimum and maximum air temperature (Tmax, Tmin), wind speed (WS), sunshine hours (SH), air pressure (PA), relative humidity (RH) as selected input data and evaporation data from pan (EP) to The output data was considered. 70% of data is used at the education level, and 30 % of the data is used at the test level. Models used with explanation coefficient evaluation (R2) Root of Mean Squares Error (RMSE) and Mean Absolute Error (MAE). The results for the two Shahroud and Mayamey stations showed that the above three models' operations are rather appropriate.

Keywords: pan evaporation, intelligent methods, shahroud, mayamey

Procedia PDF Downloads 52

Authors: Tsedeke Lambore Gemecho, Ayele Taye Goshu

Abstract:

The main objective of this study is to identify the determinants of the number of international migrants in a household and to compare regression models for count response. This study is done by collecting data from total of 2288 household heads of 16 randomly sampled districts in Hadiya and Kembata-Tembaro zones of Southern Ethiopia. The Poisson mixed models, as special cases of the generalized linear mixed model, is explored to determine effects of the predictors: age of household head, farm land size, and household size. Two ethnicities Hadiya and Kembata are included in the final model as dummy variables. Stepwise variable selection has indentified four predictors: age of head, farm land size, family size and dummy variable ethnic2 (0=other, 1=Kembata). These predictors are significant at 5% significance level with count response number of migrant. The Poisson mixed model consisting of the four predictors with random effects districts. Area specific random effects are significant with the variance of about 0.5105 and standard deviation of 0.7145. The results show that the number of migrant increases with heads age, family size, and farm land size. In conclusion, there is a significantly high number of international migration per household in the area. Age of household head, family size, and farm land size are determinants that increase the number of international migrant in households. Community-based intervention is needed so as to monitor and regulate the international migration for the benefits of the society.

Keywords: Poisson regression, GLM, number of migrant, Hadiya and Kembata Tembaro zones

Procedia PDF Downloads 265

Authors: Reza Siyami, Bahram Mirshekari

Abstract:

Seed and oil yield of rapeseed is considerably affected by weeds interference including mustard (Sinapis arvensis L.), lambsquarters (Chenopodium album L.) and redroot pigweed (Amaranthus retroflexus L.) throughout the East Azerbaijan province in Iran. To formulate the relationship between four independent growth variables measured in our experiment with a dependent variable, multiple regression analysis was carried out for the weed leaves number per plant (X1), green cover percentage (X2), LAI (X3) and leaf area per plant (X4) as independent variables and rapeseed oil yield as a dependent variable. The multiple regression equation is shown as follows: Seed essential oil yield (kg/ha) = 0.156 + 0.0325 (X1) + 0.0489 (X2) + 0.0415 (X3) + 0.133 (X4). Furthermore, the stepwise regression analysis was also carried out for the data obtained to test the significance of the independent variables affecting the oil yield as a dependent variable. The resulted stepwise regression equation is shown as follows: Oil yield = 4.42 + 0.0841 (X2) + 0.0801 (X3); R2 = 81.5. The stepwise regression analysis verified that the green cover percentage and LAI of weed had a marked increasing effect on the oil yield of rapeseed.

Keywords: green cover percentage, independent variable, interference, regression

Procedia PDF Downloads 390

Authors: Yungtai Lo

Abstract:

Two-part random effects models have been used to fit semi-continuous longitudinal data where the response variable has a point mass at 0 and a continuous right-skewed distribution for positive values. We review methods proposed in the literature for analyzing data with excess zeros. A two-part logit-log-normal random effects model, a two-part logit-truncated normal random effects model, a two-part logit-gamma random effects model, and a two-part logit-skew normal random effects model were used to examine effects of a bottle-weaning intervention on reducing bottle use and daily milk intake from bottles in toddlers aged 11 to 13 months in a randomized controlled trial. We show in all four two-part models that the intervention promoted bottle-weaning and reduced daily milk intake from bottles in toddlers drinking from a bottle. We also show that there are no differences in model fit using either the logit link function or the probit link function for modeling the probability of bottle-weaning in all four models. Furthermore, prediction accuracy of the logit or probit link function is not sensitive to the distribution assumption on daily milk intake from bottles in toddlers not off bottles.

Keywords: two-part model, semi-continuous variable, truncated normal, gamma regression, skew normal, Pearson residual, receiver operating characteristic curve

Procedia PDF Downloads 326

Authors: Suraj Mehrotra

Abstract:

The stock market is a volatile, bustling marketplace that is a cornerstone of economics. It defines whether companies are successful or in spiral. A thorough understanding of it is important - many companies have whole divisions dedicated to analysis of both their stock and of rivaling companies. Linking the world of finance and artificial intelligence (AI), especially the stock market, has been a relatively recent development. Predicting how stocks will do considering all external factors and previous data has always been a human task. With the help of AI, however, machine learning models can help us make more complete predictions in financial trends. Taking a look at the stock market specifically, predicting the open, closing, high, and low prices for the next day is very hard to do. Machine learning makes this task a lot easier. A model that builds upon itself that takes in external factors as weights can predict trends far into the future. When used effectively, new doors can be opened up in the business and finance world, and companies can make better and more complete decisions. This paper explores the various techniques used in the prediction of stock prices, from traditional statistical methods to deep learning and neural networks based approaches, among other methods. It provides a detailed analysis of the techniques and also explores the challenges in predictive analysis. For the accuracy of the testing set, taking a look at four different models - linear regression, neural network, decision tree, and naïve Bayes - on the different stocks, Apple, Google, Tesla, Amazon, United Healthcare, Exxon Mobil, J.P. Morgan & Chase, and Johnson & Johnson, the naïve Bayes model and linear regression models worked best. For the testing set, the naïve Bayes model had the highest accuracy along with the linear regression model, followed by the neural network model and then the decision tree model. The training set had similar results except for the fact that the decision tree model was perfect with complete accuracy in its predictions, which makes sense. This means that the decision tree model likely overfitted the training set when used for the testing set.

Keywords: machine learning, testing set, artificial intelligence, stock analysis

Procedia PDF Downloads 65

Authors: Ghada Abo-Zaid

Abstract:

Introduction: Oil is one of the most income source in Bahrain. Low oil price influence on the economy growth and the investment rate in Bahrain. For example, the economic growth was 3.7% in 2012, and it reduced to 2.9% in 2015. Investment rate was 9.8% in 2012, and it is reduced to be 5.9% and -12.1% in 2014 and 2015, respectively. The inflation rate is increased to the peak point in 2013 with 3.3 %. Objectives: The objectives here are to build statistical models to examine the effect of the interest rate inflation rate on the growth economy in Bahrain from 2000 to 2018. Methods: This study based on 18 years, and the multiple regression model is used for the analysis. All of the missing data are omitted from the analysis. Results: Regression model is used to examine the association between the Growth national product (GNP), the inflation rate, and real interest rate. We found that (i) Increase the real interest rate decrease the GNP. (ii) Increase the inflation rate does not effect on the growth economy in Bahrain since the average of the inflation rate was almost 2%, and this is considered as a low percentage. Conclusion: There is a positive impact of the real interest rate on the GNP in Bahrain. While the inflation rate does not show any negative influence on the GNP as the inflation rate was not large enough to effect negatively on the economy growth rate in Bahrain.

Keywords: growth national product, egypt, regression model, interest rate

Procedia PDF Downloads 129

Authors: Serkan Aras, Emrah Gulay

Abstract:

Combining various forecasting models is an important tool for researchers to attain more accurate forecasts. A great number of papers have shown that selecting single models as dissimilar models, or methods based on different information as possible leads to better forecasting performances. However, there is not a certain rule regarding the number of single models to be used in any combining methods. This study focuses on determining the optimal or near optimal number for single models with the help of statistical tests. An extensive experiment is carried out by utilizing some well-known time series data sets from diverse fields. Furthermore, many rival forecasting methods and some of the commonly used combining methods are employed. The obtained results indicate that some statistically significant performance differences can be found regarding the number of the single models in the combining methods under investigation.

Keywords: combined forecast, forecasting, M-competition, time series

Procedia PDF Downloads 329

Authors: Yanqi Yang, Chongshan Liao, Cheuk Hin Ho, Susan Bridges

Abstract:

Aim: To investigate student’s perceptions of using e-models in an inquiry-based curriculum. Approach: 52 second-year dental students completed a pre- and post-test questionnaire relating to their perceptions of e-models and their use in inquiry-based learning. The pre-test occurred prior to any learning with e-models. The follow-up survey was conducted after one year's experience of using e-models. Results: There was no significant difference between the two sets of questionnaires regarding student’s perceptions of the usefulness of e-models and their willingness to use e-models in future inquiry-based learning. Most of the students preferred using both plaster models and e-models in tandem. Conclusion: Students did not change their attitude towards e-models and most of them agreed or were neutral that e-models are useful in inquiry-based learning. Whilst recognizing the utility of 3D models for learning, student's preference for combining these with solid models has implications for the development of haptic sensibility in an operative discipline.

Keywords: e-models, inquiry-based curriculum, education, questionnaire

Procedia PDF Downloads 399

Authors: Selvakumar Ulaganathan, Ivo Couckuyt, Francesco Ferranti, Tom Dhaene, Eric Laermans

Abstract:

Variable-fidelity surrogate modelling offers an efficient way to approximate function data available in multiple degrees of accuracy each with varying computational cost. In this paper, a Kriging-based variable-fidelity surrogate modelling approach is introduced to approximate such deterministic data. Initially, individual Kriging surrogate models, which are enhanced with gradient data of different degrees of accuracy, are constructed. Then these Gradient enhanced Kriging surrogate models are strategically coupled using a recursive CoKriging formulation to provide an accurate surrogate model for the highest fidelity data. While, intuitively, gradient data is useful to enhance the accuracy of surrogate models, the primary motivation behind this work is to investigate if it is also worthwhile incorporating gradient data of varying degrees of accuracy.

Keywords: Kriging, CoKriging, Surrogate modelling, Variable- fidelity modelling, Gradients

Procedia PDF Downloads 530

Authors: Barbara Gronostajska

Abstract:

Currently, housing conditions of buildings executed in large panel technology are deteriorating. The article presents modernization solutions implemented throughout the variety of architectural activities (adding of balconies and staircases, connecting apartments) which guarantee very intriguing results that meet the needs and expectations of the modern society.

Keywords: housing estate, apartments, flats, modernization, plate blocks

Procedia PDF Downloads 456

Authors: Shuai Zhao, Shalendra D. Sharma, Jin Xu

Abstract:

This research uses advanced machine learning models to learn a number of hypotheses regarding political executives. Specifically, it analyses the impact these powerful leaders have on economic growth by using leaders’ data from the Archigos database from 1835 to the end of 2015. The data is processed by the AutoGluon, which was developed by Amazon. Automated Machine Learning (AutoML) and AutoGluon can automatically extract features from the data and then use multiple classifiers to train the data. Use a linear regression model and classification model to establish the relationship between leaders and economic growth (GDP per capita growth), and to clarify the relationship between their characteristics and economic growth from a machine learning perspective. Our work may show as a model or signal for collaboration between the fields of statistics and artificial intelligence (AI) that can light up the way for political researchers and economists.

Keywords: comparative politics, political executives, leaders’ characteristics, artificial intelligence

Procedia PDF Downloads 55

Authors: Yan Cheng, Yijun Shao, James Rudolph, Charlene R. Weir, Beth Sahlmann, Qing Zeng-Treitler

Abstract:

Supervised predictive models require labeled data for training purposes. Complete and accurate labeled data, i.e., a ‘gold standard’, is not always available, and imperfectly labeled data may need to serve as an alternative. An important question is if the accuracy of the labeled data creates a performance ceiling for the trained model. In this study, we trained several models to recognize the presence of delirium in clinical documents using data with annotations that are not completely accurate (i.e., weakly-supervised learning). In the external evaluation, the support vector machine model with a linear kernel performed best, achieving an area under the curve of 89.3% and accuracy of 88%, surpassing the 80% accuracy of the training sample. We then generated a set of simulated data and carried out a series of experiments which demonstrated that models trained on imperfect data can (but do not always) outperform the accuracy of the training data, e.g., the area under the curve for some models is higher than 80% when trained on the data with an error rate of 40%. Our experiments also showed that the error resistance of linear modeling is associated with larger sample size, error type, and linearity of the data (all p-values < 0.001). In conclusion, this study sheds light on the usefulness of imperfect data in clinical research via weakly-supervised learning.

Keywords: weakly-supervised learning, support vector machine, prediction, delirium, simulation

Procedia PDF Downloads 170

Authors: Annop Jaewisorn

Abstract:

This study explores the relationship between natural disasters and inequalities -both income and expenditure inequality- at a micro-level of Thailand as the first study of this nature for this country. The analysis uses a unique panel and remote-sensing dataset constructed for the purpose of this research. It contains provincial inequality measures and other economic and social indicators based on the Thailand Household Survey during the period between 1992 and 2019. Meanwhile, the data on natural disasters, which are remote-sensing data, are received from several official geophysical or meteorological databases. Employing a panel fixed effects, the results show that natural disasters significantly reduce household income and expenditure inequality as measured by the Gini index, implying that rich people in Thailand bear a higher cost of natural disasters when compared to poor people. The effect on income inequality is mainly driven by droughts, while the effect on expenditure inequality is mainly driven by flood events. The results are robust across heterogeneity of the samples, lagged effects, outliers, and an alternative inequality measure.

Keywords: inequality, natural disasters, remote-sensing data, Thailand

Procedia PDF Downloads 98

Authors: Piotr A. Kowalski, Kasper Sapala, Wiktor Warchalowski

Abstract:

PM10 is a suspended dust that primarily has a negative effect on the respiratory system. PM10 is responsible for attacks of coughing and wheezing, asthma or acute, violent bronchitis. Indirectly, PM10 also negatively affects the rest of the body, including increasing the risk of heart attack and stroke. Unfortunately, Poland is a country that cannot boast of good air quality, in particular, due to large PM concentration levels. Therefore, based on the dense network of Airly sensors, it was decided to deal with the problem of prediction of suspended particulate matter concentration. Due to the very complicated nature of this issue, the Machine Learning approach was used. For this purpose, Convolution Neural Network (CNN) neural networks have been adopted, these currently being the leading information processing methods in the field of computational intelligence. The aim of this research is to show the influence of particular CNN network parameters on the quality of the obtained forecast. The forecast itself is made on the basis of parameters measured by Airly sensors and is carried out for the subsequent day, hour after hour. The evaluation of learning process for the investigated models was mostly based upon the mean square error criterion; however, during the model validation, a number of other methods of quantitative evaluation were taken into account. The presented model of pollution prediction has been verified by way of real weather and air pollution data taken from the Airly sensor network. The dense and distributed network of Airly measurement devices enables access to current and archival data on air pollution, temperature, suspended particulate matter PM1.0, PM2.5, and PM10, CAQI levels, as well as atmospheric pressure and air humidity. In this investigation, PM2.5, and PM10, temperature and wind information, as well as external forecasts of temperature and wind for next 24h served as inputted data. Due to the specificity of the CNN type network, this data is transformed into tensors and then processed. This network consists of an input layer, an output layer, and many hidden layers. In the hidden layers, convolutional and pooling operations are performed. The output of this system is a vector containing 24 elements that contain prediction of PM10 concentration for the upcoming 24 hour period. Over 1000 models based on CNN methodology were tested during the study. During the research, several were selected out that give the best results, and then a comparison was made with the other models based on linear regression. The numerical tests carried out fully confirmed the positive properties of the presented method. These were carried out using real ‘big’ data. Models based on the CNN technique allow prediction of PM10 dust concentration with a much smaller mean square error than currently used methods based on linear regression. What's more, the use of neural networks increased Pearson's correlation coefficient (R²) by about 5 percent compared to the linear model. During the simulation, the R² coefficient was 0.92, 0.76, 0.75, 0.73, and 0.73 for 1st, 6th, 12th, 18th, and 24th hour of prediction respectively.

Keywords: air pollution prediction (forecasting), machine learning, regression task, convolution neural networks

Procedia PDF Downloads 110

Authors: Man-Yu Wong, Siyu Zhou, Zhiqiang Cao

Abstract:

In quality of life, health service utilization is an important determinant of medical resource expenditures on Colorectal cancer (CRC) care, a better understanding of the increased utilization of health services is essential for optimizing the allocation of healthcare resources to services and thus for enhancing the service quality, especially for high expenditure on CRC care like Hong Kong region. In assessing the association between the health-related quality of life (HRQOL) and health service utilization in patients with colorectal neoplasm, count data models can be used, which account for over dispersion or extra zero counts. In our data, the HRQOL evaluation is a self-reported measure obtained from a questionnaire completed by the patients, misreports and variations in the data are inevitable. Besides, there are more zero counts from the observed number of clinical consultations (observed frequency of zero counts = 206) than those from a Poisson distribution with mean equal to 1.33 (expected frequency of zero counts = 156). This suggests that excess of zero counts may exist. Therefore, we study tests for detecting zero-inflation in models with measurement error in covariates. Method: Under classical measurement error model, the approximate likelihood function for zero-inflation Poisson regression model can be obtained, then Approximate Maximum Likelihood Estimation(AMLE) can be derived accordingly, which is consistent and asymptotically normally distributed. By calculating score function and Fisher information based on AMLE, a score test is proposed to detect zero-inflation effect in ZIP model with measurement error. The proposed test follows asymptotically standard normal distribution under H0, and it is consistent with the test proposed for zero-inflation effect when there is no measurement error. Results: Simulation results show that empirical power of our proposed test is the highest among existing tests for zero-inflation in ZIP model with measurement error. In real data analysis, with or without considering measurement error in covariates, existing tests, and our proposed test all imply H0 should be rejected with P-value less than 0.001, i.e., zero-inflation effect is very significant, ZIP model is superior to Poisson model for analyzing this data. However, if measurement error in covariates is not considered, only one covariate is significant; if measurement error in covariates is considered, only another covariate is significant. Moreover, the direction of coefficient estimations for these two covariates is different in ZIP regression model with or without considering measurement error. Conclusion: In our study, compared to Poisson model, ZIP model should be chosen when assessing the association between condition-specific HRQOL and health service utilization in patients with colorectal neoplasm. and models taking measurement error into account will result in statistically more reliable and precise information.

Keywords: count data, measurement error, score test, zero inflation

Procedia PDF Downloads 263

Authors: Siham G. Farag

Abstract:

The main objective of this paper is to provide a new methodology for road safety assessment in Oman through the development of suitable accident prediction models. GLM technique with Poisson or NBR using SAS package was carried out to develop these models. The paper utilized the accidents data of 31 un-signalized T-intersections during three years. Five goodness-of-fit measures were used to assess the overall quality of the developed models. Two types of models were developed separately; the flow-based models including only traffic exposure functions, and the full models containing both exposure functions and other significant geometry and traffic variables. The results show that, traffic exposure functions produced much better fit to the accident data. The most effective geometric variables were major-road mean speed, minor-road 85th percentile speed, major-road lane width, distance to the nearest junction, and right-turn curb radius. The developed models can be used for intersection treatment or upgrading and specify the appropriate design parameters of T- intersections. Finally, the models presented in this thesis reflect the intersection conditions in Oman and could represent the typical conditions in several countries in the middle east area, especially gulf countries.

Keywords: accidents prediction models (APMs), generalized linear model (GLM), T-intersections, Oman

Procedia PDF Downloads 244

Authors: Shiwei Deng, Yang Bao

Abstract:

In this paper, we present a model-based regression test suite reducing approach that uses EFSM model dependence analysis and probability-driven greedy algorithm to reduce software regression test suites. The approach automatically identifies the difference between the original model and the modified model as a set of elementary model modifications. The EFSM dependence analysis is performed for each elementary modification to reduce the regression test suite, and then the probability-driven greedy algorithm is adopted to select the minimum set of test cases from the reduced regression test suite that cover all interaction patterns. Our initial experience shows that the approach may significantly reduce the size of regression test suites.

Keywords: dependence analysis, EFSM model, greedy algorithm, regression test

Procedia PDF Downloads 403

Authors: Shaan Khosla, Jon Krohn

Abstract:

In a recruiting context, machine learning models are valuable for recommendations: to predict the best candidates for a vacancy, to match the best vacancies for a candidate, and compile a set of similar candidates for any given candidate. While useful to create these models, validating their accuracy in a recommendation context is difficult due to a sparsity of data. In this report, we use network graph data to generate useful representations for candidates and vacancies. We use candidates and vacancies as network nodes and designate a bi-directional link between them based on the candidate interviewing for the vacancy. After using node2vec, the embeddings are used to construct a validation dataset with a ranked order, which will help validate new recommender systems.

Keywords: AI, machine learning, NLP, recruiting

Procedia PDF Downloads 61

Authors: Ren-Jieh Kuo, Chun-Shou Huang

Abstract:

The research of juice flavor forecasting has become more important in China. Due to the fast economic growth in China, many different kinds of juices have been introduced to the market. If a beverage company can understand their customers’ preference well, the juice can be served more attractively. Thus, this study intends to introduce the basic theory and computing process of grapes juice flavor forecasting based on support vector regression (SVR). Applying SVR, BPN and LR to forecast the flavor of grapes juice in real data, the result shows that SVR is more suitable and effective at predicting performance.

Keywords: flavor forecasting, artificial neural networks, Support Vector Regression, China

Procedia PDF Downloads 455

Authors: N. E. Gallagher, C. E. Neville, N. Lyner, J. Yarnell, C. C. Patterson, J. E. Gallacher, Y. Ben-Shlomo, A. Fehily, J. V. Woodside

Abstract:

Hearing loss is highly prevalent in older people and can reduce quality of life substantially. Emerging research suggests that potentially modifiable risk factors, including risk factors previously related to cardiovascular disease risk, may be associated with a decreased or increased incidence of hearing loss. This has prompted investigation into the possibility that certain nutrients, foods or dietary patterns may also be associated with incidence of hearing loss. The aim of this study was to determine any associations between dietary patterns and hearing loss in men enrolled in the Caerphilly study. The Caerphilly prospective cohort study began in 1979-1983 with recruitment of 2512 men aged 45-59 years. Dietary data was collected using a self-administered, semi-quantitative, 56-item food frequency questionnaire (FFQ) at baseline (1979-1983), and 7-day weighed food intake (WI) in a 30% sub-sample, while pure-tone unaided audiometric threshold was assessed at 0.5, 1, 2 and 4 kHz, between 1984 and 1988. Principal components analysis (PCA) was carried out to determine a posteriori dietary patterns and multivariate linear and logistic regression models were used to examine associations with hearing level (pure tone average (PTA) of frequencies 0.5, 1, 2 and 4 kHz in decibels (dB)) for linear regression and with hearing loss (PTA>25dB) for logistic regression. Three dietary patterns were determined using PCA on the FFQ data- Traditional, Healthy, High sugar/Alcohol avoider. After adjustment for potential confounding factors, both linear and logistic regression analyses showed a significant and inverse association between the Healthy pattern and hearing loss (P<0.001) and linear regression analysis showed a significant association between the High sugar/Alcohol avoider pattern and hearing loss (P=0.04). Three similar dietary patterns were determined using PCA on the WI data- Traditional, Healthy, High sugar/Alcohol avoider. After adjustment for potential confounding factors, logistic regression analyses showed a significant and inverse association between the Healthy pattern and hearing loss (P=0.02) and a significant association between the Traditional pattern and hearing loss (P=0.04). A Healthy dietary pattern was found to be significantly inversely associated with hearing loss in middle-aged men in the Caerphilly study. Furthermore, a High sugar/Alcohol avoider pattern (FFQ) and a Traditional pattern (WI) were associated with poorer hearing levels. Consequently, the role of dietary factors in hearing loss remains to be fully established and warrants further investigation.

Keywords: ageing, diet, dietary patterns, hearing loss

Procedia PDF Downloads 210

Authors: Stephanie Watson, Fabrice Stephenson, Conrad Pilditch, Carolyn Lundquist

Abstract:

Spatial data detailing both the distribution and density of functionally important marine species are needed to inform management decisions. Species distribution models (SDMs) have proven helpful in this regard; however, models often focus only on species occurrences derived from spatially expansive datasets and lack the resolution and detail required to inform regional management decisions. Boosted regression trees (BRT) were used to produce high-resolution SDMs (250 m) at two spatial scales predicting probability of occurrence, abundance (count per sample unit), density (count per km2) and uncertainty for seven coastal seafloor taxa that vary in habitat usage and distribution to examine prediction differences and implications for coastal management. We investigated if small scale regionally focussed models (82,000 km2) can provide improved predictions compared to data-rich national scale models (4.2 million km2). We explored the variability in predictions across model type (occurrence vs abundance) and model scale to determine if specific taxa models or model types are more robust to geographical variability. National scale occurrence models correlated well with broad-scale environmental predictors, resulting in higher AUC (Area under the receiver operating curve) and deviance explained scores; however, they tended to overpredict in the coastal environment and lacked spatially differentiated detail for some taxa. Regional models had lower overall performance, but for some taxa, spatial predictions were more differentiated at a localised ecological scale. National density models were often spatially refined and highlighted areas of ecological relevance producing more useful outputs than regional-scale models. The utility of a two-scale approach aids the selection of the most optimal combination of models to create a spatially informative density model, as results contrasted for specific taxa between model type and scale. However, it is vital that robust predictions of occurrence and abundance are generated as inputs for the combined density model as areas that do not spatially align between models can be discarded. This study demonstrates the variability in SDM outputs created over different geographical scales and highlights implications and opportunities for managers utilising these tools for regional conservation, particularly in data-limited environments.

Keywords: Benthic ecology, spatial modelling, multi-scalar modelling, marine conservation.

Procedia PDF Downloads 42

Authors: Carol Anne Hargreaves

Abstract:

A key issue in stock investment is how to select representative features for stock selection. The objective of this paper is to firstly determine whether an automated stock investment system, using machine learning techniques, may be used to identify a portfolio of growth stocks that are highly likely to provide returns better than the stock market index. The second objective is to identify the technical features that best characterize whether a stock’s price is likely to go up and to identify the most important factors and their contribution to predicting the likelihood of the stock price going up. Unsupervised machine learning techniques, such as cluster analysis, were applied to the stock data to identify a cluster of stocks that was likely to go up in price – portfolio 1. Next, the principal component analysis technique was used to select stocks that were rated high on component one and component two – portfolio 2. Thirdly, a supervised machine learning technique, the logistic regression method, was used to select stocks with a high probability of their price going up – portfolio 3. The predictive models were validated with metrics such as, sensitivity (recall), specificity and overall accuracy for all models. All accuracy measures were above 70%. All portfolios outperformed the market by more than eight times. The top three stocks were selected for each of the three stock portfolios and traded in the market for one month. After one month the return for each stock portfolio was computed and compared with the stock market index returns. The returns for all three stock portfolios was 23.87% for the principal component analysis stock portfolio, 11.65% for the logistic regression portfolio and 8.88% for the K-means cluster portfolio while the stock market performance was 0.38%. This study confirms that an automated stock investment system using machine learning techniques can identify top performing stock portfolios that outperform the stock market.

Keywords: machine learning, stock market trading, logistic regression, cluster analysis, factor analysis, decision trees, neural networks, automated stock investment system

Procedia PDF Downloads 131

Authors: D.Mendes, J. Henriques, P. Carvalho, T. Rocha, S. Paredes, R. Cabiddu, R. Trimer, R. Mendes, A. Borghi-Silva, L. Kaminsky, E. Ashley, R. Arena, J. Myers

Abstract:

The continued exploration of clinically relevant predictive models continues to be an important pursuit. Cardiorespiratory fitness (CRF) portends clinical vital information and as such its accurate prediction is of high importance. Therefore, the aim of the current study was to develop a data-driven model, based on computational intelligence techniques and, in particular, clustering approaches, to predict CRF. Two prediction models were implemented and compared: 1) the traditional Wasserman/Hansen Equations; and 2) an interpretable clustering approach. Data used for this analysis were from the 'FRIEND - Fitness Registry and the Importance of Exercise: The National Data Base'; in the present study a subset of 10690 apparently healthy individuals were utilized. The accuracy of the models was performed through the computation of sensitivity, specificity, and geometric mean values. The results show the superiority of the clustering approach in the accurate estimation of CRF (i.e., maximal oxygen consumption).

Keywords: cardiorespiratory fitness, data-driven models, knowledge extraction, machine learning

Procedia PDF Downloads 262

Authors: Paul Gurr, Maik Moser

Abstract:

In Europe, some countries recently abolished air passenger taxes (APT), while others issued or consider issuing an APT. From a fiscal perspective, APT can benefit the environment, while generating a vast amount of tax revenue with relatively low administration costs. However, they may have significant negative effects on the economy. Focusing on the German air passenger tax issued 2011, this work estimates the elasticity of tourism outflows using data on passenger departures from German airports between 2010 and 2016 aggregated by destination country. The results are obtained by estimating a model of the demand for outbound tourism. In line with theory, the regression results indicate a negative relationship between taxes and departures from Germany. Furthermore, on average, an increase of the air passenger tax rate results in a relatively higher decrease of passenger departures. The elasticity of tourism outflows can be used to estimate tax revenue changes and hence evaluate possible policy actions. Neglecting environmental reasons, the results suggest that tax revenue might be maximized by reducing the air passenger tax rate. Besides Germany, this work is also important for countries which have or consider implementing APT.

Keywords: air passenger tax, Germany, Outbound tourism, panel data

Procedia PDF Downloads 272