scholarly journals Using structural break inference for forecasting time series

2021 ◽  
Author(s):  
Gantungalag Altansukh ◽  
Denise R. Osborn
Keyword(s):  
Time Series ◽  
Monte Carlo ◽  
Confidence Interval ◽  
Likelihood Ratio ◽  
Productivity Growth ◽  
Structural Break ◽  
Forecast Accuracy ◽  
Point Estimate ◽  
Confidence Set ◽  
Explicit Consideration

AbstractRather than relying on a potentially poor point estimate of a coefficient break date when forecasting, this paper proposes averaging forecasts over sub-samples indicated by a confidence interval or set for the break date. Further, we examine whether explicit consideration of a possible variance break and the use of a two-step methodology improves forecast accuracy compared with using heteroskedasticity robust inference. Our Monte Carlo results and empirical application to US productivity growth show that averaging using the likelihood ratio-based confidence set typically performs well in comparison with other methods, while two-step inference is particularly useful when a variance break occurs concurrently with or after any coefficient break.

scholarly journals Optimal State Space Reconstruction via Monte Carlo Decision Tree Search

2021 ◽  
Author(s):  
Kai Hauke Kraemer ◽  
Maximilian Gelbrecht ◽  
Induja Pavithran ◽  
R.I. Sujith ◽  
Norbert Marwan
Keyword(s):  
Time Series ◽  
Monte Carlo ◽  
Time Delay ◽  
Objective Function ◽  
State Space ◽  
Multivariate Time Series ◽  
Forecast Accuracy ◽  
Recurrence Plot ◽  
Optimal Time ◽  
State Space Reconstruction

Abstract A novel idea for an optimal time delay state space reconstruction from uni- and multivariate time series is presented. The entire embedding process is considered as a game, in which each move corresponds to an embedding cycle and is subject to an evaluation through an objective function. This way the embedding procedure can be modeled as a tree, in which each leaf holds a specific value of the objective function. By using a Monte Carlo ansatz the proposed algorithm populates the tree with many leafs by computing different possible embedding paths and the final embedding is chosen as that particular path, which ends at the leaf with the lowest achieved value of the objective function. The method aims to prevent getting stuck in a local minimum of the objective function and can be used in a modular way, enabling practitioners to choose a statistic for possible delays in each embedding cycle as well as a suitable objective function themselves. As a proof of concept, we demonstrate the superiority of the proposed method over the classical time delay embedding methods using a variety of application examples. We compare recurrence plot based statistics inferred from reconstructions of a Lorenz-96 system and highlight an improved forecast accuracy for map-like model data as well as for paleoclimate isotope time series. Finally we utilize state space reconstruction for the detection of causality and its strength between observables of a gas turbine type thermoacoustic combustor.

A COMPARATIVE STUDY BETWEEN UNIVARIATE AND BIVARIATE TIME SERIES MODELS FOR CRUDE PALM OIL INDUSTRY IN PENINSULAR MALAYSIA

2020 ◽  
Vol 5 (1) ◽  
pp. 374
Author(s):  
Pauline Jin Wee Mah ◽  
Nur Nadhirah Nanyan
Keyword(s):  
Time Series ◽  
Palm Oil ◽  
Moving Average ◽  
Forecast Accuracy ◽  
Peninsular Malaysia ◽  
Time Series Models ◽  
Crude Palm Oil ◽  
Univariate Time Series ◽  
Import And Export ◽  
Bivariate Time Series

The main purpose of this study is to compare the performances of univariate and bivariate models on four time series variables of the crude palm oil industry in Peninsular Malaysia. The monthly data for the four variables, which are the crude palm oil production, price, import and export, were obtained from Malaysian Palm Oil Board (MPOB) and Malaysian Palm Oil Council (MPOC). In the first part of this study, univariate time series models, namely, the autoregressive integrated moving average (ARIMA), fractionally integrated autoregressive moving average (ARFIMA) and autoregressive autoregressive (ARAR) algorithm were used for modelling and forecasting purposes. Subsequently, the dependence between any two of the four variables were checked using the residuals’ sample cross correlation functions before modelling the bivariate time series. In order to model the bivariate time series and make prediction, the transfer function models were used. The forecast accuracy criteria used to evaluate the performances of the models were the mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The results of the univariate time series showed that the best model for predicting the production was ARIMA  while the ARAR algorithm were the best forecast models for predicting both the import and export of crude palm oil. However, ARIMA  appeared to be the best forecast model for price based on the MAE and MAPE values while ARFIMA  emerged the best model based on the RMSE value.  When considering bivariate time series models, the production was dependent on import while the export was dependent on either price or import. The results showed that the bivariate models had better performance compared to the univariate models for production and export of crude palm oil based on the forecast accuracy criteria used.

scholarly journals The Safety of COVID-19 Vaccinations—We Should Rethink the Policy

Vaccines ◽  
2021 ◽  
Vol 9 (7) ◽  
pp. 693
Author(s):  
Harald Walach ◽  
Rainer J. Klement ◽  
Wouter Aukema
Keyword(s):  
Side Effects ◽  
Confidence Interval ◽  
Field Study ◽  
Adverse Reactions ◽  
Safety Data ◽  
European Medicines Agency ◽  
Point Estimate ◽  
Drug Reactions ◽  
Vaccination Policy ◽  
Risks And Benefits

Background: COVID-19 vaccines have had expedited reviews without sufficient safety data. We wanted to compare risks and benefits. Method: We calculated the number needed to vaccinate (NNTV) from a large Israeli field study to prevent one death. We accessed the Adverse Drug Reactions (ADR) database of the European Medicines Agency and of the Dutch National Register (lareb.nl) to extract the number of cases reporting severe side effects and the number of cases with fatal side effects. Result: The NNTV is between 200–700 to prevent one case of COVID-19 for the mRNA vaccine marketed by Pfizer, while the NNTV to prevent one death is between 9000 and 50,000 (95% confidence interval), with 16,000 as a point estimate. The number of cases experiencing adverse reactions has been reported to be 700 per 100,000 vaccinations. Currently, we see 16 serious side effects per 100,000 vaccinations, and the number of fatal side effects is at 4.11/100,000 vaccinations. For three deaths prevented by vaccination we have to accept two inflicted by vaccination. Conclusions: This lack of clear benefit should cause governments to rethink their vaccination policy.

scholarly journals Prospective validation of Canadian TIA Score and comparison with ABCD2 and ABCD2i for subsequent stroke risk after transient ischaemic attack: multicentre prospective cohort study

BMJ ◽  
2021 ◽  
pp. n49
Author(s):  
Jeffrey J Perry ◽  
Marco L A Sivilotti ◽  
Marcel Émond ◽  
Ian G Stiell ◽  
Grant Stotts ◽  
...  
Keyword(s):  
Emergency Department ◽  
Cohort Study ◽  
Carotid Artery ◽  
Confidence Interval ◽  
Carotid Endarterectomy ◽  
Likelihood Ratio ◽  
Transient Ischaemic Attack ◽  
Carotid Artery Stenting ◽  
Stroke Risk ◽  
Ischaemic Attack

Abstract Objective To validate the previously derived Canadian TIA Score to stratify subsequent stroke risk in a new cohort of emergency department patients with transient ischaemic attack. Design Prospective cohort study. Setting 13 Canadian emergency departments over five years. Participants 7607 consecutively enrolled adult patients attending the emergency department with transient ischaemic attack or minor stroke. Main outcome measures The primary outcome was subsequent stroke or carotid endarterectomy/carotid artery stenting within seven days. The secondary outcome was subsequent stroke within seven days (with or without carotid endarterectomy/carotid artery stenting). Telephone follow-up used the validated Questionnaire for Verifying Stroke Free Status at seven and 90 days. All outcomes were adjudicated by panels of three stroke experts, blinded to the index emergency department visit. Results Of the 7607 patients, 108 (1.4%) had a subsequent stroke within seven days, 83 (1.1%) had carotid endarterectomy/carotid artery stenting within seven days, and nine had both. The Canadian TIA Score stratified the risk of stroke, carotid endarterectomy/carotid artery stenting, or both within seven days as low (risk ≤0.5%; interval likelihood ratio 0.20, 95% confidence interval 0.09 to 0.44), medium (risk 2.3%; interval likelihood ratio 0.94, 0.85 to 1.04), and high (risk 5.9% interval likelihood ratio 2.56, 2.02 to 3.25) more accurately (area under the curve 0.70, 95% confidence interval 0.66 to 0.73) than did the ABCD2 (0.60, 0.55 to 0.64) or ABCD2i (0.64, 0.59 to 0.68). Results were similar for subsequent stroke regardless of carotid endarterectomy/carotid artery stenting within seven days. Conclusion The Canadian TIA Score stratifies patients’ seven day risk for stroke, with or without carotid endarterectomy/carotid artery stenting, and is now ready for clinical use. Incorporating this validated risk estimate into management plans should improve early decision making at the index emergency visit regarding benefits of hospital admission, timing of investigations, and prioritisation of specialist referral.

A Radiated-Noise Simulator for Underwater Target Based on Vector Time Series

2014 ◽  
Vol 644-650 ◽  
pp. 4023-4026
Author(s):  
Yang Ju ◽  
Xin Yong Wang
Keyword(s):  
Time Series ◽  
Confidence Interval ◽  
Time Series Model ◽  
Experimental Results ◽  
Small Probability ◽  
Radiated Noise ◽  
True Value ◽  
Underwater Target ◽  
Vector Time Series

The vector time series model for simulating the underwater target radiated-noise is developed in this paper. Experimental results show that the true value lying outside the confidence interval would be a small probability event.

An Enquiry Into Linear Conflict Models

2021 ◽  
Author(s):  
Klaus B. Beckmann ◽  
Lennart Reimer
Keyword(s):  
Time Series ◽  
Monte Carlo ◽  
Bounded Rationality ◽  
Dynamic Models ◽  
Conflict Analysis ◽  
Monte Carlo Techniques ◽  
Linear Dynamic ◽  
Rich Variety ◽  
Time Series Analyses ◽  
Linear Dynamic Models

This monograph generalises, and extends, the classic dynamic models in conflict analysis (Lanchester 1916, Richardson 1919, Boulding 1962). Restrictions on parameters are relaxed to account for alliances and for peacekeeping. Incrementalist as well as stochastic versions of the model are reviewed. These extensions allow for a rich variety of patterns of dynamic conflict. Using Monte Carlo techniques as well as time series analyses based on GDELT data (for the Ethiopian-Eritreian war, 1998–2000), we also assess the empirical usefulness of the model. It turns out that linear dynamic models capture selected phases of the conflict quite well, offering a potential taxonomy for conflict dynamics. We also discuss a method for introducing a modicum of (bounded) rationality into models from this tradition.

Sign in / Sign up

Export Citation Format

Share Document