Direct Estimation Research Articles

Bond pairs and the term structure

AbstractIn the US Treasury bond market, the existence of a bond pair (two bonds with the same maturity but different coupons) is shown to allow the computation of the zero‐coupon interest rate for that maturity directly from the bond prices, as well as the zero‐coupon interest rates for adjacent maturity bonds with the same number of coupon payments. Since the 2008–2009 financial crisis, the number of bond pairs has increased, allowing for the direct estimation from bond prices of the zero‐coupon interest rates for an average of 180 individual maturities for bond maturities between 6 months and 30 years. The bond pairs approach outperforms popular yield‐curve‐fitting models in accurately reproducing original bond prices.

Distilling dynamical knowledge from stochastic reaction networks

Stochastic reaction networks are widely used in the modeling of stochastic systems across diverse domains such as biology, chemistry, physics, and ecology. However, the comprehension of the dynamic behaviors inherent in stochastic reaction networks is a formidable undertaking, primarily due to the exponential growth in the number of possible states or trajectories as the state space dimension increases. In this study, we introduce a knowledge distillation method based on reinforcement learning principles, aimed at compressing the dynamical knowledge encoded in stochastic reaction networks into a singular neural network construct. The trained neural network possesses the capability to accurately predict the state conditional joint probability distribution that corresponds to the given query contexts, when prompted with rate parameters, initial conditions, and time values. This obviates the need to track the dynamical process, enabling the direct estimation of normalized state and trajectory probabilities, without necessitating the integration over the complete state space. By applying our method to representative examples, we have observed a high degree of accuracy in both multimodal and high-dimensional systems. Additionally, the trained neural network can serve as a foundational model for developing efficient algorithms for parameter inference and trajectory ensemble generation. These results collectively underscore the efficacy of our approach as a universal means of distilling knowledge from stochastic reaction networks. Importantly, our methodology also spotlights the potential utility in harnessing a singular, pretrained, large-scale model to encapsulate the solution space underpinning a wide spectrum of stochastic dynamical systems.

Direct Amortized Likelihood Ratio Estimation

We introduce a new amortized likelihood ratio estimator for likelihood-free simulation-based inference (SBI). Our estimator is simple to train and estimates the likelihood ratio using a single forward pass of the neural estimator. Our approach directly computes the likelihood ratio between two competing parameter sets which is different from the previous approach of comparing two neural network output values. We refer to our model as the direct neural ratio estimator (DNRE). As part of introducing the DNRE, we derive a corresponding Monte Carlo estimate of the posterior. We benchmark our new ratio estimator and compare to previous ratio estimators in the literature. We show that our new ratio estimator often outperforms these previous approaches. As a further contribution, we introduce a new derivative estimator for likelihood ratio estimators that enables us to compare likelihood-free Hamiltonian Monte Carlo (HMC) with random-walk Metropolis-Hastings (MH). We show that HMC is equally competitive, which has not been previously shown. Finally, we include a novel real-world application of SBI by using our neural ratio estimator to design a quadcopter. Code is available at https://github.com/SRI-CSL/dnre.

Association of autoimmune and allergic diseases with senile cataract: a bidirectional two-sample Mendelian randomization study.

Many observational studies have been reported that patients with autoimmune or allergic diseases seem to have a higher risk of developing senile cataract, but the views are not consistent. In order to minimize the influence of reverse causality and potential confounding factors, we performed Mendelian Randomization (MR) analysis to investigate the genetic causal associations between autoimmune, allergic diseases and senile cataract. Single nucleotide polymorphisms associated with ten common autoimmune and allergic diseases were obtained from the IEU Open genome-wide association studies (GWAS) database. Summary-level GWAS statistics for clinically diagnosed senile cataract were obtained from the FinnGen research project GWAS, which consisted of 59,522 individuals with senile cataracts and 312,864 control individuals. MR analysis was conducted using mainly inverse variance weighted (IVW) method and further sensitivity analysis was performed to test robustness. As for ten diseases, IVW results confirmed that type 1 diabetes (OR = 1.06; 95% CI = 1.05-1.08; p = 2.24×10-12), rheumatoid arthritis (OR = 1.05; 95% CI = 1.02-1.08; p = 1.83×10-4), hypothyroidism (OR = 2.4; 95% CI = 1.42-4.06; p = 1.12×10-3), systemic lupus erythematosus (OR = 1.02; 95% CI = 1.01-1.03; p = 2.27×10-3), asthma (OR = 1.02; 95% CI = 1.01-1.03; p = 1.2×10-3) and allergic rhinitis (OR = 1.07; 95% CI = 1.02-1.11; p = 2.15×10-3) were correlated with the risk of senile cataract. Celiac disease (OR = 1.04; 95% CI = 1.01-1.08; P = 0.0437) and atopic dermatitis (OR = 1.05; 95% CI = 1.01-1.10; P = 0.0426) exhibited a suggestive connection with senile cataract after Bonferroni correction. These associations are consistent across weighted median and MR Egger methods, with similar causal estimates in direction and magnitude. Sensitivity analysis further proved that these associations were reliable. The results of the MR analysis showed that there were causal relationships between type 1 diabetes, rheumatoid arthritis, hypothyroidism, systemic lupus erythematosus, asthma, allergic rhinitis and senile cataract. To clarify the possible role of autoimmune and allergy in the pathophysiology of senile cataract, further studies are needed.

Assessing the Accuracy of Systolic Aortic Pressure Estimation From a Brachial Cuff Alone: A Comparison With Radial Tonometry.

A novel method for estimating central systolic aortic pressure (cSAP) has emerged, relying solely on the peripheral mean (MBP) and diastolic (DBP) blood pressures. We aimed to assess the accuracy of this Direct Central Blood Pressure estimation using cuff alone (DCBPcuff = MBP2/DBP) in comparison to the use of a generalized transfer function to derive cSAP from radial tonometry (cSAPtono). This retrospective analysis involved the International Database of Central Arterial Properties for Risk Stratification (IDCARS) data (Aparicio et al., Am J Hypertens 2022). The dataset encompassed 10,930 subjects from 13 longitudinal cohort studies worldwide (54.8% women; median age 46.0 years; office hypertension: 40.1%; treated: 61.0%), documenting cSAPtono via SphygmoCor calibrated against brachial systolic BP (SBP) and DBP. Our analysis focused on aggregate group data from 12/13 studies (89% patients) where a full BP dataset was available. A 35% form factor was used to estimate MBP = (DBP + (0.35 × (SBP-DBP)), from which DCBPcuff was derived. The predefined acceptable error for cSAPtono estimation was set at ≤ 5 mm Hg. The cSAPtono values ranged from 103.8-127.0 mm Hg (n = 12). The error between DCBPcuff and cSAPtono was 0.2 ± 1.4 mm Hg, with no influence of the mean. Errors ranged from -1.8 to 2.9 mm Hg across studies. No significant difference in errors was observed between BP measurements obtained via oscillometry (n = 9) vs. auscultation (n = 3) (P = 0.50). Using published aggregate group data and a 35% form factor, DCBPcuff demonstrated remarkable accuracy in estimating cSAPtono, regardless of the BP measurement technique. However, given that individual BP values were unavailable, further documentation is required to establish DCBPcuff's precision.

Blind instrument response function identification from fluorescence decays

Time-resolved fluorescence spectroscopy plays a crucial role when studying dynamic properties of complex photochemical systems. Nevertheless, the analysis of measured time decays and the extraction of exponential lifetimes often requires either the experimental assessment or the modeling of the instrument response function (IRF). However, the intrinsic nature of the IRF in the measurement process, which may vary across measurements due to chemical and instrumental factors, jeopardizes the results obtained by reconvolution approaches. In this paper, we introduce a novel methodology, called blind instrument response function identification (BIRFI), which enables the direct estimation of the IRF from the collected data. It capitalizes on the properties of single exponential signals to transform a deconvolution problem into a well-posed system identification problem. To delve into the specifics, we provide a step-by-step description of the BIRFI method and a protocol for its application to fluorescence decays. The performance of BIRFI is evaluated using simulated and time-correlated single-photon counting data. Our results demonstrate that the BIRFI methodology allows an accurate recovery of the IRF, yielding comparable or even superior results compared with those obtained with experimental IRFs when they are used for reconvolution by parametric model fitting.

A unified expression for estimating Poisson’s ratio from a hinged–hinged beam with a use for abnormality detection

Cycle fatigue weakens a structure. Poisson’s ratio is a sensitive indicator of cycle fatigue. The amount of fatigue is proportional to the severity of an abnormality, such as a crack, delamination or other imperfection. Our prior work confirms that direct estimates of Poisson’s ratio by Cornu’s method must be identical for all antinodes of a mode shape of a homogeneous, thin, hinged–hinged beam. Therefore, if a direct estimate at any antinode was found to be significantly different from the other estimates at the other antinodes, then that would simultaneously indicate fatigue and identify the location of the abnormality. We use simulations of Besselograms and apply Cornu’s method to directly estimate Poisson’s ratio at all antinodes of a beam. We transform finite element simulations of the out-of-plane bending displacements of mode shapes by simulating time-average scanning digital holography. We show that amount of disagreement between the direct estimation Poisson’s ratio and the true value of Poisson’s ratio depends on the ratio of the distance between nodes of a mode shape divided by the width of the beam. We call this distance between the nodes of a mode shape the ‘Span’. We then define a unified expression for predicting the simulated direct estimate given the true value of Poisson’s ratio and the Span-to-Width ratio. Next, we invert the unified expression, which enables us to use the Span-to-Width ratio and simulated direct estimates of Poisson’s ratio as the independent variables to compute the adjusted value of Poisson’s ratio. We find that our inverse expression makes it possible to obtain an estimate of Poisson’s ratio that is within ±2% of the true value for thin hinged–hinged beams and plates that have a Span-to-Width ratio greater than 1. We also find that if the material is known to have a Poisson’s ratio greater than zero, which can be assessed by our previous statement, then we can extend our confidence to Span-to-Width ratios greater than 0.1. That was the lower limit of our testing. Finally, we show how to use direct estimates of Poisson’s ratio to statistically detect and localize an abnormality along a hinged–hinged plate or beam. We believe that our combination of Cornu’s method and time-averaging could eventually offer the ability to directly estimate Poisson’s ratio from hinged–hinged beams and plates as well as to unambiguously monitor cycle fatigue for changes in effective material properties, predict time-to-failure, and do so without contact.

A Note on Simultaneous Confidence Intervals for Direct, Indirect and Synthetic Estimators

Direct, indirect and synthetic estimators have a long history in official statistics. While model-based or model-assisted approaches have become very popular, direct and indirect estimators remain the predominant standard and are therefore important tools in practice. This is mainly due to their simplicity, including low data requirements, assumptions and straightforward inference. With the increasing use of domain estimates in policy, the demands on these tools have also increased. Today, they are frequently used for comparative statistics. This requires appropriate tools for simultaneous inference. We study devices for constructing simultaneous confidence intervals and show that simple tools like the Bonferroni correction can easily fail. In contrast, uniform inference based on max-type statistics in combination with bootstrap methods, appropriate for finite populations, work reasonably well. We illustrate our methods with frequently applied estimators of totals and means.

Some summability methods for Dunkl–Gamma‐type operators including Appell polynomials

In this paper, we give some properties of the Dunkl–Gamma‐type operators including Appell polynomials, using into consideration the generalized power summability method. In the first section are given moments of the new defined operators. In second section are given some direct estimation related to the Dunkl–Gamma‐type operators, including Korovkin‐type theorem. In the third section, we give some results related to the weighted spaces of continuous functions, and in the last section, we give some properties in the sense of ‐statistically convergence, including Voronovskaya and Grüss–Voronovskaya‐type theorem.

Developing a New ANN Model to Estimate Daily Actual Evapotranspiration Using Limited Climatic Data and Remote Sensing Techniques for Sustainable Water Management

Accurate estimations of actual evapotranspiration (ETa) are essential to various environmental issues. Artificial intelligence-based models are a promising alternative to the most common direct ETa estimation techniques and indirect methods by remote sensing (RS)-based surface energy balance models. Artificial Neural Networks (ANNs) are proven to be suitable for predicting reference evapotranspiration (ETo) and ETa based on RS data. This study aims to develop a methodology based on ANNs for estimating daily ETa values using NDVI and land surface temperature, coupled with limited site-specific climatic variables in a large irrigation catchment. The ANN model has been applied to the two different scenarios. Data from only the 38 days of satellite overpass dates was selected in Scenario I, while in Scenario II all datasets, i.e., the 769-day data were used. An irrigation scheme, located in the Mediterranean region of Turkiye, was selected, and a total of 38 Landsat images and local climatic data collected in 2021 and 2022 were used in the ANN model. The ETa results by the ANN model for Scenarios I and II showed that the R2 values for training (0.79 and 0.86), testing (0.75 and 0.81), and the entire dataset (0.76 and 0.84) were all remarkably high. Moreover, the results of the new ANN model in two scenarios showed an acceptable agreement with ETa-METRIC values. The proposed ANN model demonstrated the potential for obtaining daily ETa using limited climatic data and RS imagery. As a result, the suggested ANN model for daily ETa computation offers a trustworthy way to determine crop water usage in real time for sustainable water management in agriculture. It may also be used to assess how crop evapotranspiration in drought-prone areas will be affected by climate change in the 21st century.

Seasonal patterns of phytoplankton photoacclimation in a dimictic oligotrophic lake over 5 yr: Implications for chlorophyll as an estimate of biomass

AbstractChlorophyll a is commonly used as a surrogate for algal biomass in lakes and oceans even though phytoplankton can readily adjust intracellular chlorophyll concentration to changes in the environment through a cellular process termed photoacclimation. Marine studies have documented fluctuations in cellular chlorophyll to year‐round changes in light intensity and nutrients using the chlorophyll : carbon ratio to estimate cellular pigment density. However, similar freshwater studies are almost exclusively limited to chlorophyll : carbon ratios at the summertime deep chlorophyll maximum (DCM) depth in less productive water. This study expands direct estimates of freshwater photoacclimation to all seasons of the year utilizing optically derived estimates of algal carbon and chlorophyll fluorescence from high‐frequency (daily) data. We describe patterns of photoacclimation at daily, seasonal, and interannual time scales over a 5‐yr period. We show that increases in chlorophyll density at the cellular level not only occurred at the DCM depth during summer stratification but were a dominant and recurring feature when the lake was vertically mixing in fall, winter, and spring. Pigment density within the mixed layer was negatively correlated with light level, indicating a photoacclimation driven response to reduced light intensity. Predicting how cellular pigment density is affected by the surrounding environment in a replicated range of lake systems is needed if chlorophyll is to be accurately used as an estimate of algal biomass in freshwaters.

Intersectional inequities and longitudinal prevalence estimates of opioid use disorder in Massachusetts 2014–2020: a multi-sample capture-recapture analysis

As overdoses continue to increase worldwide, accurate estimates are needed to understand the size of the population at risk and address health disparities. Capture-recapture methods may be used in place of direct estimation at nearly any geographic level (e.g., city, state, country) to estimate the size of the population with opioid use disorder (OUD). We performed a multi-sample capture-recapture analysis with persons aged 18-64 years to estimate the prevalence of OUD in Massachusetts from 2014 to 2020, stratified by sex and race/ethnicity. We used seven statewide administrative data sources linked at the individual level. We developed log-linear models to estimate the unknown OUD-affected population. Uncertainty was characterized using 95% confidence intervals (95% CI) on the total counts and prevalence estimates. The estimated OUD prevalence increased from 5.47% (95% CI=4.89%, 5.98%) in 2014 to 5.79% (95% CI=5.34%, 6.19%) in 2020. Prevalence among Hispanic females doubled (2.46% in 2014 to 4.23% in 2020) and prevalence rose to nearly 10% among Black non-Hispanic males and Hispanic males from 2014 through 2019. Estimates for Black non-Hispanic females more than doubled from 2014 through 2019 (3.39% to 7.09%), and then decreased to 5.69% in2020. This study is the first to provide OUD prevalence trend estimates by binary sex and race/ethnicity at a state level using capture-recapture methods. Using these methods as the international overdose crisis worsens can allow jurisdictions to appropriately allocate resources and targeted interventions to marginalised populations. NIDA.

GNSS direct position estimation-inspired positioning with pseudorange correlogram for urban navigation

Multipath (MP) reception has been among the main issues for accurate and reliable positioning in urban environments. It has been shown to introduce positioning errors of up to tens of meters for conventional two-step (2SP) receivers. The direct position estimation (DPE) has been introduced as a more robust positioning algorithm compared to the conventional two-step (2SP) receivers in terms of MP mitigation. However, its high computational load prevents DPE from real-time positioning for commercial receivers. Thus, we present a novel grid-based maximum likelihood estimation (MLE) algorithm based on DPE by making use of pseudorange measurements to obtain the correlogram on a predefined searching space. Unlike DPE, which performs correlations at the intermediate frequency (IF) level, correlations are done by directly comparing the code phase of each candidate position, velocity, and timing with the incoming pseudorange. This way, the proposed method retains MP mitigation properties from DPE through the use of MLE from DPE and allows for a significantly reduced computational load compared to DPE. The proposed method was tested with both open-sourced datasets collected in urban environments as well as IF simulation data, and its performance is evaluated against a 2SP receiver. Results show that the proposed method manages to acquire the MP mitigation capability of DPE and outperforms 2SP by up to around 23% in the tested urban datasets and 91% in the simulation data, at a much-reduced computational time. The resilience of our proposed method against MP and NLOS could even potentially offer applications in geodetic networks, where robust estimators are traditionally employed to counteract outliers.

Probabilistic 3D motion model for object tracking in aerial applications

AbstractVisual object tracking, crucial in aerial applications such as surveillance, cinematography, and chasing, faces challenges despite AI advancements. Current solutions lack full reliability, leading to common tracking failures in the presence of fast motions or long‐term occlusions of the subject. To tackle this issue, a 3D motion model is proposed that employs camera/vehicle states to locate a subject in the inertial coordinates. Next, a probability distribution is generated over future trajectories and they are sampled using a Monte Carlo technique to provide search regions that are fed into an online appearance learning process. This 3D motion model incorporates machine‐learning approaches for direct range estimation from monocular images. The model adapts computationally by adjusting search areas based on tracking confidence. It is integrated into DiMP, an online and deep learning‐based appearance model. The resulting tracker is evaluated on the VIOT dataset with sequences of both images and camera states, achieving a 68.9% tracking precision compared to DiMP's 49.7%. This approach demonstrates increased tracking duration, improved recovery after occlusions, and faster motions. Additionally, this strategy outperforms random searches by about 3.0%.

New formulation for predicting total dissolved gas supersaturation in dam reservoir: application of hybrid artificial intelligence models based on multiple signal decomposition

Total dissolved gas (TDG) concentration plays an important role in the control of the aquatic life. Elevated TDG can cause gas-bubble trauma in fish (GBT). Therefore, controlling TDG fluctuation has become of great importance for different disciplines of surface water environmental engineering.. Nowadays, direct estimation of TDG is expensive and time-consuming. Hence, this work proposes a new modelling framework for predicting TDG based on the integration of machine learning (ML) models and multiresolution signal decomposition. The proposed ML models were trained and validated using hourly data obtained from four stations at the United States Geological Survey. The dataset are composed from: (i) water temperature (Tw), (ii) barometric pressure (BP), and (iii) discharge (Q), which were used as the input variables for TDG prediction. The modelling strategy is conducted based on two different steps. First, six singles ML model namely: (i) multilayer perceptron neural network, (ii) Gaussian process regression, (iii) random forest regression, (iv) random vector functional link, (v) adaptive boosting, and (vi) Bootstrap aggregating (Bagging), were developed for predicting TDG using Tw, BP, and Q, and their performances were compared. Second, a new framework was introduced based on the combination of empirical mode decomposition (EMD), the variational mode decomposition (VMD), and the empirical wavelet transform (EWT) preprocessing signal decomposition algorithms with ML models for building new hybrid ML models. Hence, the Tw, BP, and Q signals were decomposed to extract the intrinsic mode functions (IMFs) by using the EMD and VMD methods and the multiresolution analysis (MRA) components by using the EWT method. Then after, the IMFs and MRA components were selected and regraded as new input variables for the ML models and used as an integral part thereof. The single and hybrid prediction models were compared using several statistical metrics namely, root mean square error, mean absolute error, coefficient of determination (R2), and Nash–Sutcliffe efficiency (NSE). The single and hybrid models were trained several times with high number of repetitions, depending on the kind of modeling process. The obtained results using single models gave good agreement between the predicted TDG and the situ measured dataset. Overall, the Bagging model performed better than the other five models with R2 and NSE values of 0.906 and 0.902, respectively. However, the extracted IMFs and MRA components using the EMD, VMD and the EWT have contributed to an improvement of the hybrid models’ performances, for which the R2 and NSE were significantly increased reaching the values of 0.996 and 0.995. Experimental results showed the superiority of hybrid models and more importantly the importance of signal decomposition in improving the predictive accuracy of TDG.Graphical abstract

Mendelian randomization study shows no causal relationship between psychiatric disorders and glaucoma in European and East Asian populations.

Background: Glaucoma is a leading cause of blindness strongly associated with psychiatric disorders, but the causal association between glaucoma and psychiatric disorders remains uncertain because of the susceptibility of observational studies to confounding and reverse causation. This study aims to explore the potential causal association between glaucoma and three highly related psychiatric disorders (Depression, Insomnia, and Schizophrenia) in the European and East Asian populations using a two-sample Mendelian randomization analysis. Methods: Instrumental variables (IVs) of depression, insomnia, and schizophrenia in the European population were obtained after strict filtering. Summary-level data for glaucoma and glaucoma subtypes (primary open-angle glaucoma and primary closed-angle glaucoma) were obtained as outcomes. The inverse variance weighting (IVW) method was used as the primary method. Additionally, the causal effect was evaluated in the East Asian population using the same methods to validate analysis results. The robustness of these results was confirmed using heterogeneity, pleiotropy, and Steiger directionality test. Results: The primary MR results indicated that genetically driven psychiatric disorders were not causally associated with glaucoma (Depression: odds ratio (OR): 1.15, 95% confidence interval (CI): 0.93-1.42, p = 0.20; Insomnia: OR: 1.14, 95% CI: 0.63-2.05, p = 0.66; Schizophrenia: OR: 1.00, 95% CI: 0.93-1.08, p = 0.95), either with the risk of glaucoma subtypes in the European population. Meanwhile, results in the East Asian population were consistent with the results among the European population (Depression: OR = 1.38, CI 0.75-2.53, p = 0.30; Insomnia: OR = 0.99, CI 0.83-1.18, p = 0.93; Schizophrenia: OR = 1.06, CI 0.94-1.20, p = 0.34) with similar causal estimates in direction. Consistency was obtained by corroborating with other supporting methods. Besides, the robustness of the results was proved and the directionality test confirmed our estimation of potential causal direction (p < 0.001). Conclusion: This study found a non-causal association between psychiatric disorders and the risk of glaucoma in the European and East Asian populations, which contradicts many existing observational reports, indicating that increased psychiatric disorders in glaucoma patients were more likely modifiable rather not inheritable.

Application of Liquid Chromatography Coupled to Mass Spectrometry for Direct Estimation of the Total Levels of Adenosine and Its Catabolites in Human Blood.

Adenosine is a multifunctional nucleoside with several roles across various levels in organisms. Beyond its intracellular involvement in cellular metabolism, extracellular adenosine potently influences both physiological and pathological processes. In relation to its blood level, adenosine impacts the cardiovascular system, such as heart beat rate and vasodilation. To exploit the adenosine levels in the blood, we employed the liquid chromatography method coupled with mass spectrometry (LC-MS). Immediately after collection, a blood sample mixed with acetonitrile solution that is either enriched with 13C-labeled adenosine or a newly generated mixture is transferred into the tubes containing the defined amount of 13C-labeled adenosine. The 13C-enriched isotopic adenosine is used as an internal standard, allowing for more accurate quantification of adenosine. This novel protocol for LC-MS-based estimation of adenosine delivers a rapid, highly sensitive, and reproducible means for quantitative estimation of total adenosine in blood. The method also allows for quantification of a few catabolites of adenosine, i.e., inosine, hypoxanthine, and xanthine. Our current setup did not allow for the detection or quantifying of uric acid, which is the final product of adenosine catabolism. This advancement provides an analytical tool that has the potential to enhance our understanding of adenosine's systemic impact and pave the way for further investigations into its intricate regulatory mechanisms.

Composite estimators for domain estimation and sensitivity performance interval of their weights

Some composite estimators based on various combinations of two different existing estimators are obtained for domain estimation. The estimation of weights and thus obtaining optimum weights to combine two or more different existing direct and indirect estimators to form composite estimators are not an easy task for practitioners due to many reasons. To account for the absence of optimum weights, we obtained the sensitivity performance intervals for weights with respect to the proposed composite estimator. Subsequently, we determined the sensible values of the involved weights. The aim of this procedure was to confine the superiority for different composite combinations i.e., simple direct vs. direct ratio, simple direct vs. synthetic ratio and direct ratio vs. synthetic ratio composite estimators as compared to the existing estimators.

Assessment of color changes and adverse effects of over-the-counter bleaching protocols: a systematic review and network meta-analysis.

To assess color change efficacy and the adverse effects of varied over-the-counter (OTC) bleaching protocols. The study included randomized clinical trials evaluating color changes from OTC bleaching agents. Nine databases were searched, including the partial capture of the grey literature. The RoB2 tool analyzed the individual risk of bias in the studies. Frequentist network meta-analyses compared treatments through common comparators (∆Eab* and ∆SGU color changes, and tooth sensitivity), integrating direct and indirect estimates and using the mean and risk differences as effect measures with respective 95% confidence intervals. The GRADE approach assessed the certainty of the evidence. Overall, 37 remaining studies constituted the qualitative analysis, and ten composed the meta-analyses. The total sample included 1,932 individuals. ∆Eab* was significantly higher in groups 6% hydrogen peroxide (HP) strips (≥ 14h). ∆SGU was significantly higher in groups at-home 10% carbamide peroxide (CP) (≥ 14h), followed by 6% HP strips (≥ 14h) and 3% HP strips (≥ 14h). At-home 10% CP (7-13h) and placebo showed lower risks of tooth sensitivity without significant differences between these treatments. Considering the low level of evidence, OTC products presented satisfactory short-term effects on tooth bleaching compared to the placebo, with little to no impact on dentin hypersensitivity and gingival irritation. OTC products are proving to be practical alternatives for tooth whitening. However, patients should be advised about the possible risks of carrying out such procedures without professional supervision.

SMALL AREA ESTIMATION WITH HIERARCHICAL BAYES FOR CROSS-SECTIONAL AND TIME SERIES SKEWED DATA

Small Area Estimation (SAE) is a method based on modeling for estimating small area parameters, that applies Linear Mixed Model (LMM) as its basic. It is conventionally solved with Empirical Best Linear Unbiased Prediction (EBLUP). The main requirement for LMM to produce high precision estimates is normally distributed. The observation unit is food crop farmer households from Sulawesi Tenggara Province to estimate food and non-food per capita expenditure at the district/city level using SAE that has been positively skewed. Applying EBLUP for positively skewed data will result less accurate estimates. Meanwhile, transformation will be potentially result biased estimates. Therefore, the problem of skewed data and small area level in this research was completed by Hierarchical Bayes (HB) on combination cross-sectional and time series under skew-normal distribution assumption. The results obtained were skew-normal SAE HB model was significantly reducing Relative Root Mean Squared Error (RRMSE) than the direct estimation. It indicates that SAE modeling is able to provide a shrinkage effect on the direct estimation results. But, there is slightly different interpretating between direct estimation and skew-normal SAE HB. It is possible because the modeling used assumption that the autocorrelation coefficient is equal to 1 or known as the random walk effect. However, in reality, Susenas is not a panel data, so unit of observation for each time period may be different. Therefore, further research should be compared it with the skew-normal or another skewed distribution that assumes the autocorrelation coefficient is unknown and should be estimated in the model.