Prediction Of Future Values Research Articles

Simulations and predictions of future values in the time-homogeneous load-sharing model

Simulations and predictions of future values in the time-homogeneous load-sharing model

Flexible modeling of multiple nonlinear longitudinal trajectories with censored and non-ignorable missing outcomes.

Multivariate nonlinear mixed-effects models (MNLMMs) have become a promising tool for analyzing multi-outcome longitudinal data following nonlinear trajectory patterns. However, such a classical analysis can be challenging due to censorship induced by detection limits of the quantification assay or non-response occurring when participants missed scheduled visits intermittently or discontinued participation. This article proposes an extension of the MNLMM approach, called the MNLMM-CM, by taking the censored and non-ignorable missing responses into account simultaneously. The non-ignorable missingness is described by the selection-modeling factorization to tackle the missing not at random mechanism. A Monte Carlo expectation conditional maximization algorithm coupled with the first-order Taylor approximation is developed for parameter estimation. The techniques for the calculation of standard errors of fixed effects, estimation of unobservable random effects, imputation of censored and missing responses and prediction of future values are also provided. The proposed methodology is motivated and illustrated by the analysis of a clinical HIV/AIDS dataset with censored RNA viral loads and the presence of missing CD4 and CD8 cell counts. The superiority of our method on the provision of more adequate estimation is validated by a simulation study.

Assessment of Future Pattern of Rainfall in Different Zones of Kerala Using Incorporation of SARIMA, ANN and Hybrid SARIMA-ANN Models

Agriculture production all over the world is directly influenced by rainfall which is an important weather parameter. The changes in rainfall can cause the failure of crops and even lead to starvation and ruin the economy of a country. The economy of a country becomes inferior due to catastrophic circumstances like floods, drought, and landslides. Thus the prediction of future values with the highest accuracy is very crucial to regulate and avoiding the undesirable influences of instabilities in rainfall. Under current research work, SARIMA, ANN, and hybrid SARIMA-ANN models are applied to identify the future pattern and for availing essential proposals for scheduling agriculture procedures such that variations in rainfall may not affect the economy. The data of monthly rainfall was collected from RARS, Pilicode for the northern zone, RARS, Pattambi for the central zone, and RARS, Vellayani for the southern zone. The results revealed that ANN model predicted future values of rainfall with the highest precision for the northern and central zone, whereas for the southern zone, SARIMA (1,0,1) (2,0,0)12 gave anticipated values with more accuracy. The comparison of projected rainfall among different zones indicated that the northern part might receive the highest amount of rainfall, the central zone indicated moderate rainfall, and the southern part of Kerala with the least amount of rainfall. The study also recommended the farmers take necessary safeguards to regulate the adverse influences of fluctuations in rainfall such that it might not affect agricultural production and the economy of the country.

Natural Time Series Parameters Forecasting: Validation of the Pattern-Sequence-Based Forecasting (PSF) Algorithm; A New Python Package

Climate change has contributed substantially to the weather and land characteristic phenomena. Accurate time series forecasting for climate and land parameters is highly essential in the modern era for climatologists. This paper provides a brief introduction to the algorithm and its implementation in Python. The pattern-sequence-based forecasting (PSF) algorithm aims to forecast future values of a univariate time series. The algorithm is divided into two major processes: the clustering of data and prediction. The clustering part includes the selection of an optimum value for the number of clusters and labeling the time series data. The prediction part consists of the selection of a window size and the prediction of future values with reference to past patterns. The package aims to ease the use and implementation of PSF for python users. It provides results similar to the PSF package available in R. Finally, the results of the proposed Python package are compared with results of the PSF and ARIMA methods in R. One of the issues with PSF is that the performance of forecasting result degrades if the time series has positive or negative trends. To overcome this problem difference pattern-sequence-based forecasting (DPSF) was proposed. The Python package also implements the DPSF method. In this method, the time series data are first differenced. Then, the PSF algorithm is applied to this differenced time series. Finally, the original and predicted values are restored by applying the reverse method of the differencing process. The proposed methodology is tested on several complex climate and land processes and its potential is evidenced.

Analysis and predictive validity of Kelantan River flow using RQA and time series analysis

An analysis of the Kelantan River flow for the period 2000 to 2014 and its predictive validity have been undertaken by utilizing recurrence plot (RP), recurrence quantification analysis (RQA), and auto regressive integrated moving average (ARIMA). The study presents the analysis results for flood detection and prediction of future values on a holdout sample of the 2014 data. The ARIMA model provides a better forecast when compared with the RQA predictive model. The analysis of the daily river flow dynamics reveals an abrupt change in the system and the detection of an attractor with outliers indicating higher series values; an indication of an early warning Models’ performance was based on the RMSE criterion and ARIMA model predicted a better result. The chaotic nature of the time series has also been investigated. This study could be used in understanding disastrous consequences of a river flooding.

The importance of making testable predictions: A cautionary tale.

We found a startling correlation (Pearson ρ > 0.97) between a single event in daily sea surface temperatures each spring, and peak fish egg abundance measurements the following summer, in 7 years of approximately weekly fish egg abundance data collected at Scripps Pier in La Jolla California. Even more surprising was that this event-based result persisted despite the large and variable number of fish species involved (up to 46), and the large and variable time interval between trigger and response (up to ~3 months). To mitigate potential over-fitting, we made an out-of-sample prediction beyond the publication process for the peak summer egg abundance observed at Scripps Pier in 2020 (available on bioRxiv). During peer-review, the prediction failed, and while it would be tempting to explain this away as a result of the record-breaking toxic algal bloom that occurred during the spring (9x higher concentration of dinoflagellates than ever previously recorded), a re-examination of our methodology revealed a potential source of over-fitting that had not been evaluated for robustness. This cautionary tale highlights the importance of testable true out-of-sample predictions of future values that cannot (even accidentally) be used in model fitting, and that can therefore catch model assumptions that may otherwise escape notice. We believe that this example can benefit the current push towards ecology as a predictive science and support the notion that predictions should live and die in the public domain, along with the models that made them.

Ambient Air Pollution Forecasting System using Deep Neural Networks

Air pollution is a major problem that has been recognized throughout the world. Harmful impacts of air contamination include hypersensitive reactions such as throat irritation, itchy eyes, nose, and some other serious problems. In recent years, the number of fatalities occurred due to air pollution has been increasing dramatically. In this paper, various air pollutants such as Carbon Monoxide, Methane or natural gas, LPG, and air quality at different places of city are measured using sensors. Further, the detected values are then used in the prediction of future values. The evolution of deep neural networks and Internet of Things made this possible to detect and forecast the concentration of pollutants underlying in the air. We use a special module called pyFirmata firmware which is used to connect the Arduino with python and upload the data into csv file on Jupyter Notebook. Here, the data collected is univariate i.e. it varies with only time. Though there are many statistical models to predict time series datasets such as ARIMA, their efficiency is low. Deep Neural Networks works well for predicting univariate as well as time series datasets. Hence, the Keras sequential model is employed to predict the hourly future values of air pollutants based on previous readings. The final results of prediction are compared with the actual values and error is calculated. As a result, the level of air pollutants at a particular hour can be predicted. The concentration of air pollutants in coming years, month or week helps us to reduce its concentration to lesser than the harmful or toxic range.

FORECAST OF MULTIVARIATE TIME SERIES SAMPLED FROM INDUSTRIAL MACHINERY SENSORS

Goal: To evaluate the performance of a set of forecasting methods in the prediction of future values on a dataset of time series collected from sensors installed in an industrial gas turbine.
 Design/Methodology/Approach: Forecasting methods tested include the use of multivariate and univariate neural networks (FNN and LSTM), exponential smoothing and ARIMA models.
 Results: Results show that the use of ARIMA models to forecast on the dataset is the best default method to apply, and is the only forecasting method that consistently beats a simple naïve no-change model.
 Limitation of the investigation: There was a focus on evaluating neural networks. This limited resources available to evaluate other forecasting methods. There is no guarantee that it would not be possible to find neural networks capable of yielding better forecasts than the ones achieved by the best performing methods in this research.
 Practical implications: The broadest possible implications of the results are that the best default method to forecast industrial machinery time series is the use of ARIMA models. Additionally, neural networks are not capable of beating methods well stablished within the forecasting community, namely ARIMA models.
 Originality/Value: To the best of the authors’ knowledge, there is a scarce amount of published evaluations of multiple forecasting methods on data from real machines. This knowledge is useful for the understanding of the best forecasting methods available for the estimation of machine’s RUL using sensor time series.

On model-based nowcasting for highly disaggregated levels

Nowadays, national and international organizations experience an increasing demand for timely and disaggregated socio-economic indicators. More recently, this demand extends to the request for nowcasting indicators. Small Area Estimation has a long tradition in indicator prediction for high levels of disaggregation; but when speaking of ‘prediction’, this notation refers to the fact that the centre of interest is a random parameter. Prediction of future values, or similarly, nowcasting has hardly been studied so far. Yet, mixed models based Small Area Estimation is designed for imputing (missing) values, and these models can easily account for temporal correlation. Therefore, model assisted nowcasting would be a natural extension. This article reviews existing methods under this perspective to highlight the necessary ingredients, and then propose nowcasting procedures for highly disaggregated indicators that could already be used with the today’s available software.

Forecasting Electricity Consumption Using SARIMA Method in IBM SPSS Software

Forecasting is a prediction of future values based on historical data. It can be conducted using various methods such as statistical methods or machine learning techniques. Electricity is a necessity of modern life. Hence, accurate forecasting of electricity demand is important. Overestimation will cause a waste of energy but underestimation leads to higher operation costs. Univesity Tun Hussein Onn Malaysia (UTHM) is a developing Malaysian technical university, therefore there is a need to forecast UTHM electricity consumption for future decisions on generating electric power, load switching, and infrastructure development. The monthly UTHM electricity consumption data exhibits seasonality-periodic fluctuations. Thus, the seasonal Autoregressive Integrated Moving Average (SARIMA) method was applied in IBM SPSS software to predict UTHM electricity consumption for 2019 via Box-Jenkins method and Expert Modeler. There were a total of 120 observations taken from January year 2009 to December year 2018 to build the models. The best model from both methods is SARIMA(0, 1, 1)(0, 1, 1)12. It was found that the result through the Box-Jenkins method is approximately the same with the result generated through Expert Modeler in SPSS with MAPE of 8.4%.

Tracking of Blood Pressure in Children and Adolescents in Germany in the Context of Risk Factors for Hypertension.

Blood pressure (BP) tracking from childhood to adulthood has two aspects: the ranking stability relative to others over time and the prediction of future values. This study investigates BP tracking in children and adolescents in Germany in the context of hypertension risk factors. BP was measured and analyzed in 2542 participants of the German Health Examination Survey for Children and Adolescents (t0 2003-2006; 3 to 17-year olds) and of a six year follow-up “Motorik Modul” (t1 2009-2012; 9 to 24-year olds). BP tracking coefficients were calculated from Spearman's rank-order correlations. Predictive values and logistic regression models were used to forecast t1-BP above the hypertension threshold from t0-BP as well as from baseline and follow-up hypertension risk factors. BP tracking was moderate (0.33-0.50 for SBP and 0.19-0.39 for DBP) with no statistically significant differences between sex and age groups. Baseline hypertensive BP was the strongest independent predictor of hypertensive BP at follow-up (OR 4.3 and 3.4 for age groups 3-10 and 11-17 years) after adjusting for sex, BMI trajectories, birthweight, parental hypertension, and age-group dependent-sports/physical activity. However, the positive predictive value of baseline hypertensive BP for hypertensive BP at follow-up in 3- to 10-year olds was only 39% (34% in 11- to 17-year olds) and increased only moderately in the presence of additional risk factors. Our analysis with population-based data from Germany shows that BP in children and adolescents tracks only moderately over six years. BP in childhood is the strongest independent predictor of future BP but its predictive value is limited.

Prediction of Physiological Response over Varying Forecast Lengths with a Wearable Health Monitoring Platform.

The goal of this study is to characterize the accuracy of prediction of physiological responses for varying forecast lengths using multi-modal data streams from wearable health monitoring platforms. We specifically focus on predicting breathing rate due to its significance in medical and exercise physiology research. We implement a nonlinear support vector machine regression model for accurate prediction of future values of these physiological signals with forecast windows of up to one minute long. We explore the effects of heart rate and various other sensing modalities in prediction of breathing rate. Results reveal that including other physiological responses and activity information captured by inertial measurements in the regression model improves the breathing rate prediction accuracy. We carried out experiments by collecting and analyzing physiological and activity data outside the lab using a wearable platform composed of various off-the-shelf sensors.

PSF: Introduction to R Package for Pattern Sequence Based Forecasting Algorithm

This paper discusses about an R package that implements the Pattern Sequence based Forecasting (PSF) algorithm, which was developed for univariate time series forecasting. This algorithm has been successfully applied to many different fields. The PSF algorithm consists of two major parts: clustering and prediction. The clustering part includes selection of the optimum number of clusters. It labels time series data with reference to such clusters. The prediction part includes functions like optimum window size selection for specific patterns and prediction of future values with reference to past pattern sequences. The PSF package consists of various functions to implement the PSF algorithm. It also contains a function which automates all other functions to obtain optimized prediction results. The aim of this package is to promote the PSF algorithm and to ease its implementation with minimum efforts. This paper describes all the functions in the PSF package with their syntax. It also provides a simple example of usage. Finally, the usefulness of this package is discussed by comparing it to auto.arima and ets, well-known time series forecasting functions available on CRAN repository.

A New Particle Swarm Optimization Based Stock Market Prediction Technique

Over the last years, the average person's interest in the stock market has grown dramatically. This demand has doubled with the advancement of technology that has opened in the International stock market, so that nowadays anybody can own stocks, and use many types of software to perform the aspired profit with minimum risk. Consequently, the analysis and prediction of future values and trends of the financial markets have got more attention, and due to large applications in different business transactions, stock market prediction has become a critical topic of research. In this paper, our earlier presented particle swarm optimization with center of mass technique (PSOCoM) is applied to the task of training an adaptive linear combiner to form a new stock market prediction model. This prediction model is used with some common indicators to maximize the return and minimize the risk for the stock market. The experimental results show that the proposed technique is superior than the other PSO based models according to the prediction accuracy.

A FORMALIZATION OF THE THEORY OF EXPERTONS. THEORETICAL FOUNDATIONS, PROPERTIES AND DEVELOPMENT OF SOFTWARE FOR ITS CALCULATION

JEL Classification: C02, C651.INTRODUCTIONMany economic situations that require some decision-making are unlikely to be treated with total objectivity and can be controversial. Sometimes, the poor quality of objective data means that many forecasts cannot be made satisfactorily by using only historical data from statistical series and so the compilation of expert information is more appropriate.Aggregating the opinions of a group of experts has been studied from different perspectives by Rantilla and Budescu (1999) to determine how a decision-maker might aggregate different types of entry data to generate an adequate global response. Given that the data provided by experts are based on a subjective opinion, however, it is normal to think that these data may contain some level of uncertainty. Treatment of this uncertainty from a probabilistic standpoint is a valid first option. Gerardi, McLean and Postlewaite (2009) tried to eliminate errors caused by such uncertainty using a Bayesian model.Treating uncertainty using fuzzy models is another valid option. With the birth of fuzzy logic, a new approach to information-gathering techniques also begins, in which the aggregation of information supplied by various experts, from the perspective of fuzzy logic work, leads to the emergence of new data fusion techniques. Exploring new techniques of data aggregation obtained from the fuzzy treatment of data brings us to the concept of experton (Kaufmann,1987, 1988), which allow all of the data expressed by a group of experts regarding their opinion of a certain vague characteristic to be drawn together.In the economic field, experton theory was developed by Kaufmann and Gil Aluja (1987, 1990, 1993). Its practical applications have been numerous, especially in the social sciences and particularly in the area of economics. Thus, Couturier and Fioleau (1996) applied the experton concept to financial diagnosis. Garcia and Lazzari (1996) introduced the experton to develop evaluation instruments, while Cassu, Ferrer and Bonet (2001) applied the concept to the study of growth prospects in economic sectors. Lopez and Mendana (2003) and Merigo, Gil Lafuente and Barcellos (2010) used it in decision-making problems. Gil-Lafuente and Bassa (2010) applied the experton to determine customer needs. Merigo, Casanovas and Yang (2014) used the concept in a multi-decision problem. Sirbiladze, Khutsishvili and Ghvaberidze (2014) introduce it for an investment problem. Jaile, Ferrer and Linares (2015) apply the concept of experton to construct a neural network that allows predictions of future values of an economic variable. And Chavez, Gonzalez-Santoyo, Flores B. and Flores J. J. (2015) use it for a human resource management problem.However, more attention has generally been paid to the practical construction of an experton following the initial construction proposed by Kaufmann and Gil Aluja and less to the mathematical formalization of the concept and its relationship with Φ-fuzzy sets and probalilistic fuzzy sets.This work aims to substantiate the concept by formalizing it and in doing so justify various properties of expertons and build software to calculate expertons that can be used in any field study.The concept of the experton which is built based on the aggregation of the opinion of various experts on a specific question is very closely linked to the concept of the fuzzy set, and as we will show, is defined as a certain type of probablistic fuzzy set, a concept introduced by Hirota (1977). We will therefore be using the concept of Φ-fuzzy sets introduced by Sambuc (1975) and Sanchez and Sambuc (1976) and the concept of Hirota's probabilistic fuzzy set (Hirota 1977, 1981, Hirota and Pedrycz 1983).This work has been divided as follows: there will be a first section on concepts, which will show that the concept of the experton is closely linked to the concept of fuzzy sets, since it will be defined as a certain type of probabilistic fuzzy set. …

Development and validation of an in vitro model for measurements of cervical root dentine permeability

ObjectivesThe aim of this series of studies was the development and validation of a new model for evaluation of dentinal hypersensitivity (DH) therapies.Materials and methodsRoots from extracted human teeth were sealed with a flowable composite. In the cervical area, a 3-mm-wide circular window was ground through the seal 1 mm deep into dentine. The pulp lumen was connected to a reservoir of artificial dentinal fluid (ADF) containing protein, mineral salts and methylene blue. At increased pulpal pressure, the ADF released through the said window was collected in containers each with 20 ml of physiologic saline for a consecutive series of 30-min intervals and ADF concentration (absorption) was determined photometrically. The model was verified by three experiments. In experiment 1, the lower limit of quantification (LLoQ, coefficient of variation = 20 % and difference of 5 standard deviations (SD) from blank) of ADF in physiologic saline was determined by measuring the absorption of 15 dilutions of ADF in physiologic saline (containing 0.625 ng to 12.5 μg methylene blue/ml) photometrically for ten times. In experiment 2, long-term linearity of ADF perfusion/outflow was investigated using 11 specimens. The ADF released through the window was collected in the said containers separately for each consecutive interval of 30 min for up to 240 min. Absorption was determined and analysed by linear regression over time. In experiment 3, perfusion before (2×) and after single treatment according to the following three groups was measured: BisGMA-based sealant (Seal&Protect®), an acidic fluoride solution (elmex fluid®) and control (no treatment).ResultsIn experiment 1, the LLoQ was 0.005 μg methylene blue/ml. In experiment 2, permeability was different within the specimens and decreased highly linearly with time, allowing the prediction of future values. In experiment 3, Seal&Protect® completely occluded dentinal tubules. elmex fluid® increased tubular permeability by about 30 % compared to control.ConclusionsA model comprising the use of artificial dentinal fluid was developed and validated allowing screening of therapeutic agents for the treatment of DH through reliable measurement of permeability of cervical root dentine.Clinical relevanceThe described in vitro model allows evaluation of potential agents for the treatment of DH at the clinically relevant cervical region of human teeth.

TIME SERIES FOR FAULT DETECTION IN AN INDUSTRIAL PILOT PLANT

Forecasting the evolution of industrial processes can be useful to discover faults. Several techniques based on analysis of time series are used to forecast the evolution of certain critical variables; however, the amount of variables makes difficult the analysis. In this way, the use of dimensionality reduction techniques such as the SOM (Self-Organizing Map) allows us to work with less data to determine the evolution of the process. SOM is a data mining technique widely used for supervision and monitoring. Since the SOM is projects data from a high dimensional space into a 2-D, the SOM reduces the number of variables. Thus, time series with the variables of the low dimensional projection can be created to make easier the prediction of future values in order to detect faults.

Investigation of Global Annual Temperature Anomalies and Statistical Inferences Based On It

Global warming problem earlier investigated mostly by scientists in climatology, now attract the attention of manyresearchers because a changing climate cause a great anxiety. The paper analyses the data of North and South hemisphere temperaturesvariation and North and South hemisphere temperatures 5 years average variation. The data collected by the National Aeronauticsand Space Administration since 1880 year are analyzed as time series in this paper. The ARMA models are fitted, their statisticalcharacteristics evaluated and predictions for future values have been calculated using the fitted models of one dimensional and multidimensionalautoregressive models. Calculated predicted values were compared to really observed values and the research resulted inthe best fitted autoregressive and moving average models well describing the global warming data.

Bayesian analysis of Box–Cox transformed linear mixed models with ARMA() dependence

In this paper, we present a Bayesian inference methodology for Box–Cox transformed linear mixed model with ARMA(p,q) errors using approximate Bayesian and Markov chain Monte Carlo methods. Two priors are proposed and put into comparisons in parameter estimation and prediction of future values. The advantages of Bayesian approach over maximum likelihood method are demonstrated by both real and simulated data.

Estimation and Prediction of Generalized Growth Curve with Grouping Variances in AR(q) Dependence Structure

AbstractIn this paper we consider maximum likelihood analysis of generalized growth curve model with the Box‐Cox transformation when the covariance matrix has AR(q) dependence structure with grouping variances. The covariance matrix under consideration is Σ = DσCDσ where C is the correlation matrix with stationary autoregression process of order q, q < p and Dσ is a diagonal matrix with p elements divided into g(≤ p) groups, i.e., Dσ is a function of {σ1, …, σg} and – 1 < ρ < 1 and σl, l = 1, …, g, are unknown. We consider both parameter estimation and prediction of future values. Results are illustrated with real and simulated data.