Last updated: Jan. 5, 2023, 4:20 p.m.
Speaker: 𝙍𝙚𝙜𝙞𝙣𝙖 𝙈. 𝙏𝙧𝙚𝙨𝙫𝙖𝙡𝙡𝙚𝙨
Date: Jan. 12, 2023, 4 p.m.
Venue: via Zoom
Abstract:
This paper proposes to use a spatio-temporal model with structural change for the spread of the coronavirus disease 2019 (COVID-19), showing its suitability in the pandemic experience in the Philippines, from the first recorded case in the country to the end of the fourth surge in October 2021. This model was originally proposed by Bastero and Barrios in 2011. The result uses a spatiotemporal model that incorporates the forward search algorithm and maximum likelihood estimation into the backfitting framework. The forward search algorithm is used to filter the effect of short-term structural change in the estimation of covariate and spatial parameters. The method that uses the effect of structural change offers good model fit especially in COVID-19 experience in the Philippines where structural change is encountered at different space and time. It is further explored in this paper how this spatiotemporal model may be incorporated in the compartmental mathematical models for COVID-19.
Keywords: spatio-temporal model, structural change, forward search, COVID-19
Last updated: April 18, 2023, 5:06 p.m.
Speaker: Honeylet Santos
Date: April 20, 2023, 5 p.m.
Venue: via Zoom
Abstract:
Statistical matching deals with methods of combining different data from different sources to get information on variables not observed in a single source. With the goal of estimating a Poisson regression model, this study explores statistical matching techniques and estimation procedures involving bootstrap. Simulation studies confirmed that Poisson regression imputation and MCMC imputation produce comparable results. It also showed that "bootstrap within" method performs well regardless of the matching method used.
Keywords:
statistical matching, count data, imputation, bootstrap
Last updated: April 18, 2023, 5:06 p.m.
Speaker: Stephen Jun Villejo
Date: April 27, 2023, 5 p.m.
Venue: via Zoom
Abstract: We propose a two-stage latent Gaussian model for a specific spatial misalignment problem. We use the integrated nested Laplace approximation (INLA) to perform inference. The first-stage model is based on the Bayesian melding which does data fusion by assuming a common latent field for the observed outcomes. We use the stochastic partial differential equations (SPDE) approach to efficiently estimate the spatial field. It induces a Markov structure on the field which leads to sparse precision matrices giving great computational gains. Uncertainty in the first stage is accounted for by simulating repeatedly from the posterior predictive distribution of the field. A simulation study was carried out to assess the impact of the sparsity of the data, the number of time points, and the specification of the priors in terms of the biases, RMSEs, and coverage probabilities. The results show that the parameters are generally estimated correctly, but there is difficulty in estimating the random field parameters. The method is applied to NO2 concentration and respiratory hospitalizations for the year 2017 in England. An increase in NO2 levels is significantly associated with an increase in the relative risks of the health outcome. Also, there is a strong spatial structure and a strong temporal autocorrelation in the risks.
Last updated: Aug. 3, 2023, 2:10 p.m.
Speaker: 𝙃𝙚𝙣𝙧𝙮 𝙅𝙤𝙨𝙚𝙥𝙝 𝙈. 𝙃𝙚𝙧𝙧𝙚𝙧𝙖, 𝙁𝘼𝙎𝙋
Date: Aug. 10, 2023, 4 p.m.
Venue: UPSS Lecture Hall I
Abstract: The talk discusses a roadmap to achieving financial freedom - its stages and the Philippine experience. Sources and expected amounts of retirement funds from mandated social insurance and private employers will be covered. Building an individual retirement account as early as possible will be highlighted to supplement these retirement benefits provided by law. A worksheet will be provided to participants to enable them to determine the amount they need to retire comfortably at the age they desire to retire.
Last updated: April 14, 2023, 6:21 p.m.
Speaker: 𝙎𝙝𝙞𝙧𝙡𝙚𝙚 𝙍. 𝙊𝙘𝙖𝙢𝙥𝙤
Date: Jan. 12, 2023, 5 p.m.
Venue: via Zoom
Abstract:
Many spatiotemporal processes such as epidemic data, signal spectrum, weather and environmental occurrences are often sparse. Philippine COVID-19 data are said to be underreported due to lack of mass testing, contact tracing plus there are unreported cases, thus making the data messy and sparse. Modeling sparse data has been a challenge since the sparse features increase the space and time complexity of models. The study postulated sparse spatial autoregressive and spatiotemporal models. It also investigated models linking sparse data to different covariates like the healthcare system, demographic and economic indicators, disease prevalence, vaccination, urbanity, and environmental factors. Sparsity regularization terms are incorporated to improve the predictive abilities of the spatiotemporal models. The parameters were estimated using a hybrid of Cochranne-Orcutt and backfitting algorithm with integrated cross validation of sparsity regularized objective function. Moreover, hurdle models were also formulated, and these were estimated using forward search algorithm and maximum likelihood estimation infused in the backfitting framework. The postulated models were fitted to simulated data and their predictive abilities were compared using mean square error (MSE) and mean absolute error (MAE). The postulated models were applied to daily COVID-19 cases and deaths across provinces and National Capital Region (NCR) cities from April 1, 2020 to September 15, 2021 in the Philippines. The models emphasize the importance of resources available in the local government that can boost the capabilities of the health care system. Pre-existing health conditions (co-morbidities) of the communities also determine prevalence and mortality rates of COVID-19 in the Philippines.
Keywords:
sparse, spatiotemporal model, backfitting algorithm, forward search, COVID-19
OFFICIAL STATEMENT OF THE FACULTY OF UP SCHOOL OF STATISTICS on the Proliferation of Surveys with Unclear Methodologies We, the faculty members ...
View our Most Read ArticleOne of the highlights of the 2015 UP School of Statistics Recognition Rites was the announcement of the recipients of the Best Undergraduate Student Paper Award.
View our most read journalSTAT SPEAKS
Date: August 10, 2023 4 PM
Speaker: 𝙃𝙚𝙣𝙧𝙮 𝙅𝙤𝙨𝙚𝙥𝙝 𝙈. 𝙃𝙚𝙧𝙧𝙚𝙧𝙖, 𝙁𝘼𝙎𝙋
View our Recent Event