Search

Air pollution, especially ground-level ozone, poses severe threats to human health and ecosystems. Accurate forecasting of ozone concentrations is essential for reducing its adverse effects. This study aims to use the functional time series approach to model ozone concentrations, a method less explored in the literature, and compare it with traditional time series and machine learning models. To this end, the ozone concentration hourly time series is first filtered for yearly seasonality using smoothing splines that lead us to the stochastic (residual) component. The stochastic component is modeled and forecast using a functional autoregressive model (FAR), where each daily ozone concentration profile is considered a single functional datum. For comparison purposes, different traditional and machine learning techniques, such as autoregressive integrated moving average (ARIMA), vector autoregressive (VAR), neural network autoregressive (NNAR), random forest (RF), and support vector machine (SVM), are also used to model and forecast the stochastic component. Once the forecast from the yearly seasonality component and stochastic component are obtained, both are added to obtain the final forecast. For empirical investigation, data consisting of hourly ozone measurements from Los Angeles from 2013 to 2017 are used, and one-day-ahead out-of-sample forecasts are obtained for a complete year. Based on the evaluation metrics, such as R 2 , root mean squared error (RMSE), and mean absolute error (MAE), the forecasting results indicate that the FAR outperforms the competitors in most scenarios, with the SVM model performing the least favorably across all cases. [ABSTRACT FROM AUTHOR]

The Central Bank of Turkey’s policy to decrease the nominal interest rate has caused episodes of severe fluctuations in Turkish lira exchange rates during 2022. According to these conditions, the daily return of the USD/TRY have attracted the risk-taker investors’ attention. Therefore, the uncertainty about the rates has pushed algorithmic traders toward finding the best forecasting model. While there is a growing tendency to employ sophisticated models to forecast financial time series, in most cases, simple models can provide more precise forecasts. To examine that claim, present study has utilized several models to predict daily exchange rates for a short horizon. Interestingly, the simple exponential smoothing model outperformed all other alternatives. Besides, in contrast to the initial inferences, the time series neither had structural break nor exhibited signs of the ARCH and leverage effects. Despite that behavior, there was undeniable evidence of a long-memory trend. That means the series tends to keep a movement, at least for a short period. Finally, the study concluded the simple models provide better forecasts for exchange rates than the complicated approaches.

Copyright of Baghdad Science Journal is the property of Republic of Iraq Ministry of Higher Education & Scientific Research (MOHESR) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Air pollution, especially ground-level ozone, poses severe threats to human health and ecosystems. Accurate forecasting of ozone concentrations is essential for reducing its adverse effects. This study aims to use the functional time series approach to model ozone concentrations, a method less explored in the literature, and compare it with traditional time series and machine learning models. To this end, the ozone concentration hourly time series is first filtered for yearly seasonality using smoothing splines that lead us to the stochastic (residual) component. The stochastic component is modeled and forecast using a functional autoregressive model (FAR), where each daily ozone concentration profile is considered a single functional datum. For comparison purposes, different traditional and machine learning techniques, such as autoregressive integrated moving average (ARIMA), vector autoregressive (VAR), neural network autoregressive (NNAR), random forest (RF), and support vector machine (SVM), are also used to model and forecast the stochastic component. Once the forecast from the yearly seasonality component and stochastic component are obtained, both are added to obtain the final forecast. For empirical investigation, data consisting of hourly ozone measurements from Los Angeles from 2013 to 2017 are used, and one-day-ahead out-of-sample forecasts are obtained for a complete year. Based on the evaluation metrics, such as R2, root mean squared error (RMSE), and mean absolute error (MAE), the forecasting results indicate that the FAR outperforms the competitors in most scenarios, with the SVM model performing the least favorably across all cases.

This research aims to be guided decision-makers in future planning by estimating the tendency of data consistently. In this context, it is thought that the integration of the Autoregressive Distributed Lag-ARDL models, gathering the independent factors and their past effects as well as the past trend of the dependent variable, with fuzzy regression methods, would give more realistic results. To prove the correctness of this idea, the Fuzzy-ARDL method has been proposed and tested the superiority of the research on the projection of USAs' annual oil consumption data examined by researchers previously. For this purpose, raw data of crude oil import price, population, gross national domestic production (GDP) per capita, and oil production variables, previously compiled annually, have been considered independent variables. Then the proposed model has been benchmarked with the other promising models from the fuzzy regression literature. As a result, according to various Accuracy Measures values, it has been seen that the proposed model outperforms the other promising models. [ABSTRACT FROM AUTHOR]

Accurate global horizontal irradiance (GHI) forecasting promotes power grid stability. Most of the research on solar irradiance forecasting has been based on a single-site analysis. It is crucial to explore multisite modeling to capture variations in weather conditions between various sites, thereby producing a more robust model. In this research, we propose the use of spatial regression coupled with Gaussian Process Regression (GP Spatial) and the GP Autoregressive Spatial model (GP-AR Spatial) for the prediction of GHI using data from seven radiometric stations from South Africa and one from Namibia. The results of the proposed methods were compared with a benchmark model, the Linear Spatial Temporal Regression (LSTR) model. Five validation sets each comprised of three stations were chosen. For each validation set, the remaining five stations were used for training. Based on root mean square error, the GP model gave the most accurate forecasts across the validation sets. These results were confirmed by the statistical significance tests using the Giacommini–White test. In terms of coverage probability, there was a 100% coverage on three validation sets and the other two had 97% and 99%. The GP model dominated the other two models. One of the study's contributions is using standardized forecasts and including a nonlinear trend covariate, which improved the accuracy of the forecasts. The forecasts were combined using a monotone composite quantile regression neural network and a quantile generalized additive model. This modeling framework could be useful to power utility companies in making informed decisions when planning power grid management, including large-scale solar power integration onto the power grid. [ABSTRACT FROM AUTHOR]

Nowadays, short-term traffic flow forecasting has gained increasing attention from researchers due to traffic congestion in many large and medium-sized cities that pose a serious threat to sustainable urban development. To this end, this research examines the forecasting performance of functional time series modeling to forecast traffic flow in the ultra-short term. An appealing feature of the functional approach is that unlike other methods, it provides information over the whole day, and thus, forecasts can be obtained for any time within a day. Within this approach, a Functional AutoRegressive (FAR) model is used to forecast the next-day traffic flow. For empirical analysis, the traffic flow data of Dublin airport link road, Ireland, collected at a fifteen-minute interval from 1 January 2016 to 30 April 2017, are used. The first twelve months are used for model estimation, while the remaining four months are for the one-day-ahead out-of-sample forecast. For comparison purposes, a widely used model, namely AutoRegressive Integrated Moving Average (ARIMA), is also used to obtain the forecasts. Finally, the models' performances are compared based on different accuracy statistics. The study results suggested that the functional time series model outperforms the traditional time series models. As the proposed method can produce traffic flow forecasts for the entire next day with satisfactory results, it can be used in decision making by transportation policymakers and city planners. [ABSTRACT FROM AUTHOR]

The COVID-19 epidemic has raised major issues with regard to modelling and forecasting outcomes such as cases, deaths and hospitalisations. In particular, the forecasting of area-specific counts of infectious disease poses problems when counts are changing rapidly and there are infection hotspots, as in epidemic situations. Such forecasts are of central importance for prioritizing interventions or making severity designations for different areas. In this paper, we consider different specifications of autoregressive dependence in incidence counts as these may considerably impact on adaptivity in epidemic situations. In particular, we introduce parameters to allow temporal adaptivity in autoregressive dependence. A case study considers COVID-19 data for 144 English local authorities during the UK epidemic second wave in late 2020 and early 2021, which demonstrate geographical clustering in new cases—linked to the then emergent alpha variant. The model allows for both spatial and time variation in autoregressive effects. We assess sensitivity in short-term predictions and fit to specification (spatial vs space-time autoregression, linear vs log-linear, and form of space decay), and show improved one-step ahead and in-sample prediction using space-time autoregression including temporal adaptivity. [ABSTRACT FROM AUTHOR]

Efficient modeling and forecasting of electricity prices are essential in today’s competitive electricity markets. However, price forecasting is not easy due to the specific features of the electricity price series. This study examines the performance of an ensemble-based technique for forecasting short-term electricity spot prices in the Italian electricity market (IPEX). To this end, the price time series is divided into deterministic and stochastic components. The deterministic component that includes long-term trends, annual and weekly seasonality, and bank holidays, is estimated using semi-parametric techniques. On the other hand, the stochastic component considers the short-term dynamics of the price series and is estimated by time series and various machine learning algorithms. Based on three standard accuracy measures, the results indicate that the ensemble-based model outperforms the others, while the random forest and ARMA are highly competitive.

Nowadays, time series analysis is an important challenge in engineering problems. This paper proposes a Comprehensive Learning Polynomial Autoregressive Model for predicting linear and nonlinear time series. The model is based on the autoregressive model but developed in a polynomial aspect to make it more robust and accurate. The model predicts future values by learning the weights of the weighted sum of the polynomial combination of previous data. The learning process for the hyperparameters and properties of the model in the training phase is performed by the metaheuristic optimization method. Using this model, we can predict nonlinear time series, as well as linear time series. The proposed method was implemented on eight standard stationary and non-stationary large-scale real-world datasets. It outperformed the state-of-the-art methods that use deep learning in seven time series and has better results compared to all other methods in six datasets. Experimental results show the advantage of the model's accuracy over other compared methods on various prediction tasks based on the root-mean-square error. [ABSTRACT FROM AUTHOR]

The research includes the application of a hybrid model by integrating between the Fuzzy Autoregressive Integrated Moving Average (FARIMA) model and the Conditional Autoregressive of Generalized Variance Heterogeneity (GARCH) model, using the residuals of the FARIMA model as inputs to the GARCH model on the weekly time series data of stock prices in the Iraqi Stock Exchange. A number of models were proposed and then a comparison was made between them using evaluation criteria. It was found that the FARIMA (1,1,0) - GARCH (1,1) hybrid model is the most appropriate model for analyzing the data under study and the most efficient in the accuracy of future prediction compared to the FARIMA model because it has less Values for prediction accuracy criteria (MSE, MAE, ME, RMSE, MAPE). [ABSTRACT FROM AUTHOR]

Financial analysis of the stock market using the historical data is the exigent demand in business and academia. This work explores the efficiency of three deep learning (Dl) techniques, namely Bayesian regularization (BE), Levenberg–Marquardt (lM), and scaled conjugate gradient (SCG), for training nonlinear autoregressive artificial neural networks (NARX) for predicting specifically the closing price of the Egyptian Stock Exchange indices (EGX-30, EGX-30-Capped, EGX-50-EWI, EGX-70, EGX-100, and NIlE). An empirical comparison is established among the experimented prediction models considering all techniques for the time horizon of 1 day, 3 days, 5 days, 7 days, 5 days and 30 days in advance, applying on all the datasets used in this study. For performance evaluation, statistical measures such as mean squared error (MSE) and correlation R are used. From the simulation result, it can be clearly suggested that BR outperforms other models for short-term prediction especially for 3 days ahead. On the other hand, lM generates better prediction accuracy than BR- and SCG-based models for long-term prediction, especially for 7-day prediction. [ABSTRACT FROM AUTHOR]

• This study applies the autoregressive latent trajectory model to probe longitudinal reading achievement in early elementary years and its relationships with classroom behavior and gender. • The autoregressive latent trajectory model performs better than the latent growth model and the autoregressive model. • Classroom behavior and gender status have a mix of positive and negative relationships with K-4 reading growth. Elementary students' ability to regulate classroom behavior is believed to play an important role in their reading ability. To date, there is some evidence articulating this relationship, but much of the research has not addressed how classroom behavior may predict the longitudinal reading achievement. The present study used data from the Early Childhood Longitudinal Study—Kindergarten Class of 2010–2011 to examine the associations of self-control, attentional focusing, and externalizing problem behaviors with reading development for elementary students from kindergarten through fourth grade. A combination of the latent growth model and the autoregressive model – the Autoregressive Latent Trajectory model is applied to describe children's increasing but later leveling off reading growth. Results from the Autoregressive Latent Trajectory model conditional on classroom behaviors show that they have a mix of positive and negative associations with reading growth. Further, a multigroup analysis reveals the gender differences in such associations. Implications for educators, researchers and policy makers are discussed. [ABSTRACT FROM AUTHOR]

Copyright of Economic Sciences is the property of Republic of Iraq Ministry of Higher Education & Scientific Research (MOHESR) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Highlights • Develop a new functional model to forecast day-ahead natural gas-flow in Germany. • Incorporate functional exogenous variables in the modeling. • Strong evidence that the proposed model outperforms several alternative models. • Improve forecast accuracy by up to 5-fold. • Results show that nominations and lagged 1-day flow are important predictors. Abstract Forecasting natural gas demand and supply is essential for an efficient operation of the German gas distribution system and a basis for the operational decisions of the transmission system operators. The German gas market is moving towards more short-term planning, in particular, day-ahead contracts. This increases the difficulty that the operators in the dispatching centre are facing, as well as the necessity of accurate forecasts. This paper presents a novel predictive model that provides day-ahead forecasts of the high resolution gas flow by developing a Functional AutoRegressive model with eXogenous variables (FARX). The predictive model allows the dynamic patterns of hourly gas flows to be described in a wide range of historical profiles, while also taking the relevant determinants data into account. By taking into account a richer set of information, FARX provides stronger performance in real data analysis, with both accuracy and high computational efficiency. Compared to several alternative models in out-of-sample forecasts, the proposed model can improve forecast accuracy by at least 12% and up to 5-fold for one node, 3% to 2-fold and 2-fold to 4-fold for the other two nodes. The results show that lagged 1-day gas flow and nominations are important predictors, and with their presence in the forecast model, temperature becomes insignificant for short-term predictions. [ABSTRACT FROM AUTHOR]

Natural gas is vital in the world's energy portfolio and is widely applied to power generation, urban heating, and manufacturing. Forecasting natural gas consumption with high accuracy is thus crucial in order to maintain a reliable supply for various applications. The demand for natural gas often exhibits different seasonal patterns regarding customers of different characteristics. The precision of forecasters will be vulnerably affected without carefully exploring the periodicity of usage. This paper proposes a novel method, Dual Convolution with Seasonal Decomposition Network, for natural gas consumption forecasting. The proposed method applies multiple seasonal-trend decomposition to separate time series into periodic patterns and residual components. In addition, local and global convolution are combined to predict series with significant fluctuations and poor periodicity. Simulations show that on city-level forecasting, the proposed method outperforms state-of-the-art methods in terms of overall prediction accuracy and variation sensitivity regardless of different time intervals. The performance of the method is robust to the forecasting horizon. The method can be deployed in practical circumstances to forecast the natural gas consumption of residential quarters, cities, or even countries in different time spans. • The multiple seasonal-trend decomposition is crucial for extracting regular patterns. • A global and local convolution network is effective in exploring long-short term features. • The autoregressive component improves the input sensitivity of neural networks. • The proposed method outperforms the state-of-the-art on actual data. [ABSTRACT FROM AUTHOR]

Trying to predict the future price of certain stocks, securities or indexes is quite a common goal, being motivated by different reasons and being based on various techniques. The present article has the same purpose, employing an ARIMA model, due to its proven effectiveness and success. Used data is comprised of monthly values for the mentioned index, on a four-year period, from 2010 to 2014, which lead to 60 recordings. The main steps for the analysis are identifying the model, estimating the parameters and the prediction itself. After each one of them is carefully conducted, a comparison is made: the predicted values for BET versus the real values for BET, in order to see if any resemblances exists, or if the differences are significant. Those resemblances or differences are explained, while the conclusion will highlight ARIMA's capacity or incapacity of forecasting in an accurate way, in the presented context. [ABSTRACT FROM AUTHOR]

Purpose - The purpose of this study is to ascertain whether nonlinearities could be present in electricity loads observed in subtropical environments, where none or little heating is required, and whether threshold autoregressive (TAR)-type regime switching models could be advantageous in the modeling of those loads. Design/methodology/approach - The actual observed load of a Brazilian regional electricity distributor from January 2013 to August 2012 was modeled using a popularly employed ARMA model for reference, and smooth and non-smooth TAR transition (non-linear) models were used as non-linear regime switching models. Findings - Evidence of nonlinearities were found in the load series, and evidence was also found on the intrinsic resistance of this type of models to structural breaks in the data. Additionally, to reacting well to asymmetries in the data, these models avoid the use of exogenous variables. Altogether, this could prove to be a definite advantage of the use of such model alternatives. Research limitations/implications - However, even if the present work may have been limited by the observation frequency of the available data, it appears TAR models appear to be a viable alternative to forecasting short-term electricity loads. Nonetheless, additional research is required to achieve a higher accuracy of forecast data. Practical implications - If such models can be successfully used, it will be a great advantage for electricity generators, as the computational effort involved in the use of such models is not significantly larger than regular linear ones. Originality/value - To our knowledge, this type of research has not yet been made with subtropical/ tropical electricity load data. [ABSTRACT FROM AUTHOR]

Air pollution is today a serious problem, caused mainly by human activity. Classical methods are not considered able to efficiently model complex phenomena as meteorology and air pollution because, usually, they make approximations or too rigid schematisations. Our purpose is a more flexible architecture (artificial neural network model) to implement a short-term CO2 emission forecasting tool applied to the cereal sector in Apulia region – in Southern Italy - to determine how the introduction of cultural methods with less environmental impact acts on a possible pollution reduction. [ABSTRACT FROM AUTHOR]

Reconstruction and prediction of spatiotemporal time series data has been a classic problem in structural health monitoring (SHM) in civil engineering applications. However, due to the explosive growth of sensing data, traditional time series analysis approaches fail in handling large-scale data with missing values. To this end, two autoregressive (AR) based matrix factorization (MF) methods are presented for missing sensor data imputation and structural response forecasting. The first model integrates the standard MF formulation with an innovative graph-based temporal regularizer, which can effectively model the nonlinear dynamics of SHM data and is computationally efficient, while the second approach introduces an additional AR-based matrix to better simulate the temporal factor thanks to its capability of learning the details of temporal evolution. Finally, the proposed methods are evaluated by using a field-recorded SHM dataset of a municipal concrete bridge, considering various missing scenarios (i.e., random, structured and mixed). The results demonstrate excellent performance of the methods which accurately recover missing entries in the time series and forecast future response. Additionally, the parametric analysis on model parameters indicates that reasonably higher rank and longer time lag improve the estimation accuracy while saving computational cost. • Introduced two autoregressive matrix factorization methods. • Applied to spatiotemporal structural monitoring data imputation and forecasting. • Demonstrated effectiveness on a concrete bridge with 9-month heterogeneous data. [ABSTRACT FROM AUTHOR]

This article introduces a novel framework for analyzing long-horizon forecasting of the near non-stationary AR(1) model. Consider the univariate AR(1) model, for which the I(1) framework specifies the random walk (RW) model and the VAR framework uses a simple ordinary least squares (OLS) regression to estimate the parameters of the model. It is clear that, in finite samples, RW forecasts have lower mean square error (MSE) than OLS forecasts when the autoregressive parameter is sufficiently close, but perhaps not equal, to unity. As a result, RW forecasts may be preferred even in situations where the random walk model itself represents a misspecification of the true time series process. In such cases, RW produces biased forecasts. Letting the autoregressive parameter shrink away from unity, this bias increases, driving up the error of RW forecasts. Eventually for sufficiently small values of the autoregressive parameter, the MSE of RW forecasts becomes equal to, then larger than, the MSE of OLS forecasts.

The purpose of this paper was to project international tourist arrivals in 12 major tourist destinations in South-East Asia and the Pacific to the year 2001. The SAS forecast autoregressive time series trend model was applied and 25 years (1970-94) of historical data was used to project international tourist flows in these destinations. The five years' projection (1997-2001) showed that tourist arrivals in the 12 destinations will increase from 69.9 million in 1997 to 104.6 million in the year 2001, with an overall growth rate of 49.6 per cent and an average annual growth rate of 10.6 per cent. The tremendons growth of international tourism in the region has been very much associated with intra-regional travel in the past. In the projected regional market share, intra-regional travel will still maintain its dominant position, increasing from 61 per cent of market share in 1997 to 63.5 per cent in 2001, followed by Europe, America, and Oceania. By the year 2001, Singapore, Hong Kong, Thailand, China, and South Korea will be the top five tourist destinations in the region. [ABSTRACT FROM PUBLISHER]

In this work, we investigate a probabilistic method for electricity price forecasting, which overcomes traditional ones. We start considering statistical methods for point forecast, comparing their performance in terms of efficiency, accuracy, and reliability, and we then exploit Neural Networks approaches to derive a hybrid model for probabilistic type forecasting. We show that our solution reaches the highest standard both in terms of efficiency and precision by testing its output on German electricity prices data. [ABSTRACT FROM AUTHOR]

• A model considering stochasticity instead of deterministic degradation shape. • Predict degradation by constructing segments iteratively with autoregressive nature. • Produce remaining useful life confidence intervals from bootstrap distribution. • The proposed model is applicable to two different lithium-ion batteries. • Comparison shows the proposed model can predict better when plunge is not present. While most existing degradation modeling methods for rechargeable batteries consider a deterministic degradation model such as exponential model, this paper presents a time series model for battery degradation paths resembling experimental data on cycle aging. This model is based on breaking down the degradation path into segments by fitting a multiple-change-point linear model, which accounts for the degradation structure by regressing the segment lengths and the slope changes. These two variables are modeled by two sub-models: an autoregressive model with covariates for the slope changes at the change points and a survival regression model for the segment lengths that allows for censored data caused by interruptions during battery cycling. The combined model is able to predict a full battery degradation path based on historical paths, and predict the remaining degradation path even based merely on the partial path. The proposed model can also be used to produce confidence intervals for battery's useful life by applying the method of parametric bootstrap to generate the empirical bootstrap distribution. The application of the proposed model is demonstrated with data from lithium iron phosphate and lithium nickel manganese cobalt oxide batteries. The comparison on prediction mean between proposed model, deterministic models with particle filter and recurrent neural network shows that the proposed model can make better prediction when capacity plunge is not present. The validation with simulations shows that the proposed model is reliable when complete historical paths are available as the simulation coverage rates are close to the nominal coverage rate 90%. [ABSTRACT FROM AUTHOR]

A different approach based on existing spatiotemporal models is developed in this research for forecasting the production flow rates of a group of adjacent wells in a reservoir. Unlike the previous single-well analysis techniques, this new methodology tries to forecast the flow rates of a number of producing wells located in a certain neighborhood. It also takes into account the variety of flow rate patterns that may take place during the production period of the wells. Regarding the time-series nature of flow rate data and the spatial relationships between multiple wells in a proximity, the Space-Time Auto-Regressive Moving-Average technique, abbreviated as STARMA, is borrowed to employ in this research to capture the spatial inter-relationships of the wells and model the flow rate data of multiple wells simultaneously. Partitioning the field into different groups of adjacent wells based on the similarity of flow rate patterns is also recommended using the spatial clustering algorithms as an enhancing part of the proposed methodology. The capability of the proposed approach in forecasting the flow rates of multiple wells is verified using the real data of a case study including 38 active oil producing wells located in a reservoir in South West Iran. The forecasts generated by the proposed method are finally compared with the predictions obtained by the current ARIMA techniques. Based on findings of this research, the proposed space time modeling of multiple well flow rates helps in estimating the more accurate forecasts, when the single well analysis techniques fail in predicting the real trend of future flow rates. • A new spatiotemporal model is presented to forecast multiple-well flow rates. • Proximity of wells is considered using STARMA algorithm. • Spatial clustering of wells is suggested to enhance the forecasting process. • The proposed methodology is verified using the real data of Iranian oil wells. • The results of this research are compared with previously-applied ARIMA technique. [ABSTRACT FROM AUTHOR]

When performing a time series analysis of continuous data, for example, from climate or environmental problems, the assumption that the process is Gaussian is often violated. Therefore, we introduce two non‐Gaussian autoregressive time series models that are able to fit skewed and heavy‐tailed time series data. Our two models are based on the Tukey g‐and‐h transformation. We discuss parameter estimation, order selection, and forecasting procedures for our models and examine their performances in a simulation study. We demonstrate the usefulness of our models by applying them to two sets of wind speed data. [ABSTRACT FROM AUTHOR]

We introduce a way to discretize a point process such as earthquakes into time series and propose a class of two‐part autoregressive (2PAR) models for continuous time series data with excess zeros to analyze the dependence structure in such discretized point processes. In the 2PAR models, we employ a Bernoulli variable to model the excess zeros and use autoregressive processes to describe the serial correlation. Using this class of models, we can model correlations that exist in either zeros or nonzeros in the data. We introduce a class of residual analysis to check the goodness‐of‐fit of the proposed models and a forecasting procedure using simulation to check the performance of the models. A simulation study illustrates that the estimators of the proposed models are consistent. Application to the 2010 Darfield earthquake sequence in New Zealand shows that the 2PAR models with serial correlation in both the presence probability and the earthquake stress release provide higher information gain against a reference model. [ABSTRACT FROM AUTHOR]