Structural Time Series Model Research Articles

Estimating Causal Effects in the Presence of Partial Interference Using Multivariate Bayesian Structural Time Series Models

Researchers regularly use synthetic control methods for estimating causal effects when a sub-set of units receive a single persistent treatment, and the rest are unaffected by the change. In many applications, however, units not assigned to treatment are nevertheless impacted by the intervention because of cross-unit interactions. This paper extends the synthetic control methods to accommodate partial interference, allowing interactions within predefined groups, but not between them. Focusing on a class of causal estimands that capture the effect both on the treated and control units, we develop a multivariate Bayesian structural time series model for generating synthetic controls that would have occurred in the absence of an intervention enabling us to estimate our novel effects. In a simulation study, we explore our Bayesian procedure’s empirical properties and show that it achieves good frequentists coverage even when the model is misspecified. Our work is motivated by an analysis of a marketing campaign’s effectiveness by an Italian supermarket chain that permanently reduced the price of hundreds of store-brand products. We use our new methodology to make causal statements about the impact on sales of the affected store-brands and their direct competitors. Our proposed approach is implemented in the CausalMBSTS R package.

Missing data imputation of high‐resolution temporal climate time series data

AbstractAnalysis of high‐resolution data offers greater opportunity to understand the nature of data variability, behaviours, trends and to detect small changes. Climate studies often require complete time series data which, in the presence of missing data, means imputation must be undertaken. Research on the imputation of high‐resolution temporal climate time series data is still at an early phase. In this study, multiple approaches to the imputation of missing values were evaluated, including a structural time series model with Kalman smoothing, an autoregressive integrated moving average (ARIMA) model with Kalman smoothing and multiple linear regression. The methods were applied to complete subsets of data from 12 month time series of hourly temperature, humidity and wind speed data from four locations along the coast of Western Australia. Assuming that observations were missing at random, artificial gaps of missing observations were studied using a five‐fold cross‐validation methodology with the proportion of missing data set to 10%. The techniques were compared using the pooled mean absolute error, root mean square error and symmetric mean absolute percentage error. The multiple linear regression model was generally the best model based on the pooled performance indicators, followed by the ARIMA with Kalman smoothing. However, the low error values obtained from each of the approaches suggested that the models competed closely and imputed highly plausible values. To some extent, the performance of the models varied among locations. It can be concluded that the modelling approaches studied have demonstrated suitability in imputing missing data in hourly temperature, humidity and wind speed data and are therefore recommended for application in other fields where high‐resolution data with missing values are common.

A Bayesian Structural Time Series Approach for Predicting Red Sea Temperatures

Sea surface temperature (SST) is a leading factor impacting coral reefs and causing bleaching events in the Red Sea. A long-term prediction of temperature patterns with an estimate of uncertainty is thus essential for environment management of the Red Sea ecosystem. In this work, we build a data-driven Bayesian structural time series model and show its effectiveness in predicting future SST seasons with a high accuracy, and identifying the main predictive factors of future SST variability among a large number of factors, including regional SST and large-scale climate indices. The modeling scheme proposed here applies an efficient hierarchical clustering to identify interconnected subregions that display distinct SST variability over the Red Sea, and a Markov Chain Monte Carlo algorithm to simultaneously select the main predictors while the time series model is being trained. In particular, numerical results indicate that monthly SST can be reliably predicted for five months ahead.

Relevant parameter changes in structural break models

Structural break time series models, which are commonly used in macroeconomics and finance, capture unknown structural changes by allowing for abrupt changes to model parameters. However, many specifications suffer from an over-parametrization issue, since typically all parameters have to change when a break occurs. We introduce a sparse change-point model to detect which parameters change over time. We propose a shrinkage prior distribution, which controls model parsimony by limiting the number of parameters that change from one structural break to another. We develop a Bayesian sampler for inference on the sparse change-point model. An extensive simulation study based on AR, ARMA and GARCH processes highlights the excellent performance of the sampler. We provide several empirical applications including an out-of-sample forecasting exercise showing that the Sparse change-point framework compares favorably with other recent time-varying parameter processes.

Energy efficiency and underlying carbon emission trends.

The present study aims to quantify the impact of energy efficiency on environmental quality in the top five greenhouse gas-emitting countries, namely, China, USA, India, Germany, and Japan. Based on annual data set from 1971 to 2016, the study explores the role of technological innovations to protect environmental quality by reducing CO2 emissions of these countries by employing structural time series model (STSM) approach. The results show that CO2 emissions in China, India, the USA, and Germany are growing faster than energy consumption, while in Japan the former is growing more slowly than energy consumption. The estimated carbon emission trends are upward sloping for China and India and downward sloping for Germany. However, for the USA and Japan, the underlying trend in carbon emission shows mix results. The persistently increasing slope of the carbon emission trends in China and India suggest that either energy-efficient appliances have not been used to protect the quality of the environment or any improvement in energy efficiency due to technological progress is offset by other exogenous factors.

A national intervention to reduce imaging for low back pain by general practitioners: a retrospective economic program evaluation using Medicare Benefits Schedule data

BackgroundThe overuse of diagnostic imaging for low back pain (LBP) in Australia results in unnecessary cost to the health system and, for patients, avoidable exposure to radiation. The 2013 NPS MedicineWise LBP program aimed to reduce unnecessary diagnostic imaging for non-specific acute LBP in the Australian primary care setting. The LBP program delivered referral pattern feedback, a decision support tool and patient information to 19,997 (60%) of registered Australian general practitioners (GPs). This study describes the findings from evaluation of the effectiveness of the 2013 LBP program at reducing X-ray and computed tomography (CT) scans of the lower back, and the financial costs and benefits of the program to the government funder.MethodsThe effectiveness of the 2013 LBP program was evaluated using population-based time-series analysis of administrative claims data of Medicare Benefits Schedule (MBS) funded X-ray and CT scan services of the lower back. The CT scan referral trend of non-GP health professionals was used as an observational control group in a Bayesian structural time-series model. A retrospective cost–benefit analysis and cost-effectiveness analysis was conducted using program costs from organisational records and reimbursement data from the MBS.ResultsThe 2013 NPS MedicineWise LBP program was associated with a statistically significant 10.85% relative reduction in the volume of CT scans of the lumbosacral region, equating to a cost reduction to the MBS of AUD$11,600,898. The best available estimate of program costs was AUD$141,154. Every dollar of funding spent on the 2013 LBP program saved AUD$82 of funding to the MBS for CT scan reimbursements. Therefore, from the perspective of the Australian Government Department of Health, the 2013 LBP program was cost saving. The program cost AUD$2.82 per CT scan averted in comparison to the scenario of no program. No association between the 2013 NPS MedicineWise LBP program and the volume of X-ray items on the MBS was observed.ConclusionsThe 2013 NPS MedicineWise LBP program reduced CT scan referral by GPs, in line with the program’s messages and clinical guidelines. Reducing this low-value care produced savings to the health system that exceeded the costs of program implementation.

A New Dynamic House-Price Index

Using a newly available panel data set containing property-specific, time-varying hedonic characteristics and sales prices, we develop a new dynamic house-price model that is suitable for out-of-sample forecasting applications such as mortgage valuation and bank stress-testing. The model is set up in a classical state-space framework and includes common factors that are univariate structural time series models scaled to form linear combinations specific to locations. Our common factors include trend, seasonal, and autoregressive components. The equations are linear and errors are Gaussian; however, the unbalanced nature of our panel data means that standard Kalman-filter smoothing algorithms are not suitable. Instead, we apply an alternative three-block Markov Chain Monte Carlo algorithm Strickland, Turner, Denham, and Mengersen (2009). We find significant in- and out-of-sample forecasting differences between our model and standard repeat-sales and hedonic regressions. We also test, and reject, two assumptions of repeat-sales estimators: a single common trend for all regions and a unit root in the index.

Modelling industrial energy demand in Saudi Arabia

Between 1986 and 2016, industrial energy consumption in Saudi Arabia increased by tenfold, making it one of the largest end-use sectors in the Kingdom. Despite its importance, there appear to be no published econometric studies on aggregate industrial energy demand in Saudi Arabia. We model aggregate industrial energy demand in Saudi Arabia using Harvey’s (1989) Structural Time Series Model, showing that it is both price and income inelastic, with estimated long-run elasticities of −0.34 and 0.60, respectively. The estimated underlying energy demand trend suggests improvements in energy efficiency starting from 2010.Applying decomposition analysis to the estimated econometric equation highlights the prominent roles of the activity effect (the growth in industrial value added) and the structure effect (the shift towards energy-intensive production) in driving industrial energy demand growth. Moreover, the decomposition shows how exogenous factors such as energy efficiency helped mitigate some of that growth, delivering cumulative savings of 6.8 million tonnes of oil equivalent (Mtoe) between 2010 and 2016.Saudi Arabia implemented a broad energy price reform program in 2016, which raised electricity, fuel, and water prices for households and industry. The decomposition results reveal that, holding all else constant, higher industrial energy prices in 2016 reduced the sector’s energy consumption by 6.9 %, a decrease of around 3.0 Mtoe. Saudi policymakers could therefore build on the current policy of energy price reform and energy efficiency standards to mitigate the rate of growth of industrial energy consumption, increase economic efficiency, and maintain industrial sector competitiveness.

Forecasting tourism demand: The Hamilton filter

Studies on tourism demand forecasting techniques can be classified into two main groups: quantitative and qualitative methods (Song & Li, 2008). The use of quantitative methods in tourism demand forecasting is more popular than qualitative methods (Song & Li, 2008). The types of time series models include naive, autoregressive, single exponential, moving average and historical averages (Song, Qiu, & Park, 2019). The decomposition of time series has been found to improve forecasting accuracy (Shabri, 2016; Zhang et al., 2017). The common techniques for time series decomposition in tourism forecasting are filters (Li, Wong, Song, & Witt, 2006); spectral analysis (Coshall, 2000); and empirical mode decompositions (Chen, Lai, & Yeh, 2012). Causal structural time series models perform less satisfactorily than univariate models (Turner & Witt, 2001). Li and Law (2019) examined the effectiveness of decomposed search engine data in forecasting tourism demand in Hong Kong. The proposed technique of using decomposed online search engine data was viable based on the out-of-sample forecast evaluation (Li & Law, 2019). When decomposing time series, it is important to account for seasonality as most tourism destinations are affected by seasonal patterns (Saayman & Botha, 2017; Vergori, 2012). Chen, Li, Wu, and Shen (2019) proposed a multiseries structural time series method as an alternative technique to seasonal tourism demand forecasting. The forecast evaluation support that the structural model is viable. Chu (2004) applied a cubic polynomial model to forecast tourist arrivals in Singapore. The cubic polynomial model was found to be less effective than the sine wave and ARIMA applications. The use of seasonal fractional models for modelling the seasonal component of Spanish tourism demand was pioneered by Gil-Alana, De Gracia, and Cunado (2004). The authors found that the number of foreigners and foreign guest nights exhibit seasonal long memory behaviour.

CICLOS ECONÔMICOS E CRIMES CONTRA O PATRIMÔNIO: UMA ANÁLISE QUASE-EXPERIMENTAL

Este artigo busca evidências das relações entre ciclos econômicos e crimescontra o patrimônio a partir de um quase experimento gerado pela implantação doPolo Naval no município brasileiro de Rio Grande (RS) a partir do ano de 2006. Paraesse fim, utilizam-se modelos estruturais de séries de tempo estimados através do Filtrode Kalman para decompor e identificar quebras estruturais nas séries de furtos e roubosregistrados mensalmente no município, no período compreendido entre janeiro de2002 e dezembro de 2015. Os resultados indicam a existência de quebras estruturaistanto em furtos, quanto em roubos, que podem estar associadas às mudançasgeradas na economia em decorrência da implementação do Polo Naval. O artigo trazevidências de que o início da construção da plataforma P-53 coincide com a reduçãode forma permanente de 80 furtos mensais e que a construção da plataforma P-55coincide com a redução permanente de 62 roubos mensais registrados no município.O artigo conclui que existem evidências de um efeito motivação em que a melhoranas condições econômicas do município foi capaz de gerar efeitos dissuasórios quereduziram os crimes contra o patrimônio.

Unobserved Components Model for Forecasting Sugarcane Yield in Haryana

Unlike classical regression analysis, the state space models have time-dependent parameters and provide a flexible class of dynamic and structural time series models. The unobserved component model (UCM) is a special type of state space models widely used to analyze and forecast time series. The present investigation has been carried out to study the trend of sugarcane(gur) yield in five districts (Ambala, Karnal, Panipat, Yamunanagar and Kurukshetra) of Haryana state using the unobserved component models with level, trend and irregular components. For this purpose, the time series data on sugarcane yield from 1966-67 to 2016-17 of Ambala and Karnal, 1971-72 to 2016-17 of Kurukshetra and 1980-81 to 2016-17 of Panipat and Yamunanagar districts have been used. For all the districts, the irregular component was found to be highly significant (p=0.01) while both level and trend component variances were observed non-significant. Significance analysis of the individual component(s) has also been performed for possible dropping of the level and trend components by setting their variances equal to zero. The state space models may be effectively used pertaining to Indian agriculture data, as it takes into account the time dependency of the underlying parameters which may further enhance the predictive accuracy of the most popularly used ARIMA models with parameter constancy. Moreover, the unobserved component model is capable of handling both stationary as well as non-stationary time series and thus found more suitable for sugarcane yield modeling which is a trended yield (i.e. non-stationary in nature).

Bayesian Structural Time Series and Regression Modeling for Sustainable Technology Management

Many companies take the sustainability of their technologies very seriously, because companies with sustainable technologies are better able to survive in the market. Thus, sustainable technology analysis is important issue in management of technology (MOT). In this paper, we study the management of sustainable technology (MOST). This focuses on the sustainable technology in various MOT fields. In the MOST, sustainable technology analysis is dependent on time periods. We propose a method of sustainable technology analysis using a Bayesian structural time series (BSTS) model based on time series data. In addition, we use the Bayesian regression to find the relational structure between technologies. To show the performance of our method and how the method can be applied to practical works, we carry out a case study using the patent data related to artificial intelligence technologies.

Using machine learning to quantify the impacts of genetically modified crops on US midwest corn yields

Global food security is becoming increasingly stressed by growing populations and climate change. To compensate for these stresses, crop yields must increase throughout the upcoming century. One of the more prominently featured solutions entails genetically modified crops, but their impacts on yields are contested. Here, we leverage machine learning techniques to examine the effects genetically modified crops have had on US corn yields. In particular, a principal components analysis conducted on US Midwest county yields reveals that the commercialization of genetically modified corn accentuated preexisting spatial disparities in production and explains approximately 6–12% of the region's inter-county variation in yields from 1980 to 2015. Additionally, counterfactual yield trajectories predicted by Bayesian structural time series models using non-genetically modified crops as synthetic controls suggest that the adoption of this biotechnology amounted to an approximate 13% increase in overall US corn yields from 1996 to 2015.

Is radiative forcing cointegrated with temperature? A further examination using a structural time series approach

Purpose The purpose of this paper is to re-examine the long-run relationship between radiative forcing (including emissions of carbon dioxide, sulphur oxides, methane and solar radiation) and temperatures from a structural time series modelling perspective. The authors assess whether forcing measures are cointegrated with global temperatures using the structural time series approach. Design/methodology/approach A Bayesian approach is used to obtain estimates that represent the uncertainty regarding this relationship. The estimated structural time series model enables alternative model specifications to be consistently compared by evaluating model performance. Findings The results confirm that cointegration between radiative forcing and temperatures is consistent with the data. However, the results find less support for cointegration between forcing and temperature data than found previously. Research limitations/implications Given considerable debate within the literature relating to the “best” way to statistically model this relationship and explain results arising as well as model performance, there is uncertainty regarding our understanding of this relationship and resulting policy design and implementation. There is a need for further modelling and use of more data. Practical implications There is divergence of views as to how best to statistically capture, explain and model this relationship. Researchers should avoid being too strident in their claims about model performance and better appreciate the role of uncertainty. Originality/value The results of this study make a contribution to the literature by employing a theoretically motivated framework in which a number of plausible alternatives are considered in detail, as opposed to simply employing a standard cointegration framework.

An unobserved component analysis of Malaysia’s house prices

PurposeThis paper aims to investigate Malaysia’s house prices behaviour by decomposing trend, cycle and stochastic component.Design/methodology/approachThe authors perform an unobserved component model of a structural time series and Markov switching model that covers the period 1999Q1 to 2015Q4.FindingsThe results reveal that the variation in house price in Malaysia is best explained by its trend level, with a small role played by the cycle component; this implies the potential for gaining returns on investments in property by investors and households. The results also show that Malaysia’s HPI cycle is between 8 and 17 years which, in relative terms, is twice the length of the growth cycle and the business cycle in the economy. Meanwhile, the overall movement of HPI is forecast to have a marginal price increase up to 2028Q2.Originality/valueAs house prices remained elevated during the year, the house price dynamic is pivotal for understanding the source of changes in house price. With major findings centred on the relationship between house prices and macroeconomic as well as policy variables, little attention has been paid to composing the trend, cycle and seasonal pattern from the house price index, thus understanding the behaviour of house prices’ unobserved components.

Assessing the effect of advertising expenditures upon sales: A Bayesian structural time series model

AbstractWe propose a robust implementation of the Nerlove‐Arrow model using a Bayesian structural time series model to explain the relationship between advertising expenditures of a countrywide fast‐food franchise network with its weekly sales. Due to the flexibility and modularity of the model, it is well suited to generalization to other markets or situations. Its Bayesian nature facilitates incorporating a priori information reflecting the manager's views, which can be updated with relevant data. This aspect of the model will be used to support the decision of the manager on the budget scheduling of the advertising firm across time and channels.

Design and development of interdisciplinary learning in a new oncology unit of study in the new integrated pharmacy curriculum

Since the first oil crisis in the 1970s, demand for energy has undergone significant transformations. This paper intends to model the demand by the transportation sector for the petroleum products in Iran. We propose a structural time-series model with an underlying trend. After transforming the model to state-space form, we apply the Kalman Filter algorithm to estimate the parameters of the model for the period between 1980 and 2016. Results indicate that the underlying trend for gasoline and diesel demand by the transportation sector is non-linear. This trend can be used as proxy applicable for studying those variables which could enable us to explain technological improvements and changes in consumer preferences for gasoline and diesel over time. Results also show that technological improvements and changes in consumers’ behavior led to a decrease in the consumption of diesel in recent years while it has no decreasing effect on gasoline. Demand functions show that gasoline and diesel are inelastic. Gasoline price elasticity of demand ranges between −0.17 and −0.24. The price elasticity of demand for diesel is very low and statistically insignificant at the 5% level. The income elasticity of demand for gasoline and diesel ranges between 0.34 and 0.48, and between 0.33 and 0.52 respectively, being strongly statistically significant. Finally, this paper proposes that the price of fuel should be determined by market mechanism to reduce fuel consumption.

Impact of a public transit strike on public bicycle share use: An interrupted time series natural experiment study

IntroductionPromoting active transportation is an important public health objective. Limited research has examined the potential of interventions that highly constrain transportation and their potential impact on cycling. From November 1-7th, 2016, Philadelphia's transit workers went on strike, stopping all transit services in the city. MethodsWe used the strike event as a natural experiment to examine the impact of public transit strikes on use of Philadelphia's bicycle share program. We estimated the impact of the strike using two separate approaches, interrupted time series and Bayesian structural time series models. We estimated the impact of the intervention overall and stratified by membership type (members and non-members). Models controlled for the weather in Philadelphia (daily temperature and precipitation), and the rate of bicycle share use per 100,000 people in Washington, Boston, and Chicago. ResultsWe estimate the strike caused an increase of between 86 and 92 trips per 100,000 population (57% increase in use) on average in Philadelphia during the strike period. After the strike ridership quickly returned to baseline, decreasing by 80 trips per 100,000 population after the strike. Similarly, members and non-member ridership increased by 41 and 49 trips per 100,000 population on average during the strike period and quickly returned to baseline, respectively. ConclusionsOur results suggest that interventions that highly constrain transit can increase active transportation but the behavior may not be sustained after transit becomes available again.

International Real Estate Review

Regional interdependence among the real estate markets in the Gulf Cooperation Council (GCC) is tested by using a variety of techniques. Econometric tests that involve error correction, symmetric/asymmetric autoregressive distributed lag (ARDL) and structural time series models are utilized. The results reveal the absence of long-run relationships, thus indicating cross-sectional efficiency. However, strong evidence is found for both short and long-run dynamic interdependence when the model allows for asymmetric responses. Finally, the results from the structural time series modeling show that a weak form of interdependence is present, which partly shows that other factors of significant impact explain for the real estate fluctuation other than the corresponding prices of the neighboring countries. Plausible fiscal and monetary policy recommendations are presented.

Influence of seasonality, salinity and temperature on catch trend of Penaeus indicus H. Milne-Edwards, 1837 in a coastal lagoon, India

The catch pattern of P. indicus in coastal lagoons is influenced by seasonal changes in physicochemical parameters of the lagoon ecosystem. In this study the effects of seasonality, salinity and water emperature of lagoon on P. indicus catch were analysed using Structural Time Series Model (STSM) and ARIMAX (Auto Regressive Integrated Moving Average with explanatory variables) modeling approach using monthly time series catch, salinity and water temperature data of the Chilika Lagoon (a Ramsar site) in India for the period from 2001 to 2015. Results showed a significant (p<0.05) increasing stochastic upward trend and two seasonal cycles for P. indicus catch in the lagoon. Salinity was found to have significant positive influence (p<0.05) and temperature to have insignificant positive influence on P. indicus catch in the lagoon.