scholarly journals Inference Based on the Best-Fitting Model Can Contribute to the Replication Crisis: Assessing Model Selection Uncertainty Using a Bootstrap Approach

2016 ◽  
Vol 23 (4) ◽  
pp. 479-490 ◽  
Author(s):  
Gitta H. Lubke ◽  
Ian Campbell
Keyword(s):  
Model Selection ◽  
Bootstrap Approach ◽  
Replication Crisis ◽  
Model Selection Uncertainty ◽  
Best Fitting ◽  
Fitting Model

scholarly journals Modeling Count Data for Healthcare Utilization: an Empirical Study of Outpatient Visits Among Vietnamese Older People

2020 ◽  
Author(s):  
Dung Duc Le ◽  
Roberto Leon-Gonzalez ◽  
Joseph Upile Matola
Keyword(s):  
Model Selection ◽  
Older People ◽  
Healthcare Utilization ◽  
Count Data ◽  
Aging Process ◽  
Cross Validation ◽  
Outpatient Visits ◽  
Best Fitting ◽  
Fold Cross Validation ◽  
Fitting Model

Abstract Background Vietnam is undergoing an unprecedented pace of aging process and is expected to experience the fastest aging process in region. Association between increasing age and health deterioration has been well-documented across settings. Consequently, demand for healthcare utilization is rising among older people. However, healthcare utilization, here measured as count data, creates challenges for modeling because such data typically has distributions that are skewed with a large mass at zero. This study compares empirical econometric strategies for the modeling of healthcare utilization (measured as the number of outpatient visits in the last 12 months), and identifies the determinants of healthcare utilization among Vietnamese older people based on the best-fitting model identified. Methods Using the Vietnam Household Living Standard Survey in 2006 (N=2426), nine econometric regression models for count data were examined to identify the best-fitting one. We used model selection criteria; statistical tests; and goodness-of-fit for in-sample model selection. In addition, we conducted 10-fold cross-validation checks to examine reliability of in-sample model selection. Finally, we utilized marginal effects to identify the factors associated with number of outpatient visits among Vietnamese older people based on the best-fitting model identified. Results We found strong evidence in favor of hurdle negative binomial model 2 (HNB2) for both in-sample selection and 10-fold cross-validation checks. The marginal effect results of the HNB2 showed that predisposing, enabling, need, and lifestyle factors were significantly associated with number of outpatient visits. The predicted probabilities for each count event showed the distinct trends of healthcare utilization among specific groups: at low count events, women and people in younger age group used more healthcare utilization than did men and their counterparts in older age groups, but a reversed trend was found at higher count events. Conclusions The findings here suggest that the HNB2 model should be considered for use in modeling counts of healthcare use. This study’s findings lay the groundwork for future research on the modeling of healthcare utilization in developing countries and those findings could be used to forecast on healthcare demand and making provisions for healthcare costs.

scholarly journals Modeling Count Data for Healthcare Utilization: an Empirical Study of Outpatient Visits Among Vietnamese Older People

2020 ◽  
Author(s):  
Dung Duc Le ◽  
Roberto Leon-Gonzalez ◽  
Joseph Upile Matola
Keyword(s):  
Model Selection ◽  
Older People ◽  
Healthcare Utilization ◽  
Count Data ◽  
Aging Process ◽  
Cross Validation ◽  
Outpatient Visits ◽  
Best Fitting ◽  
Fold Cross Validation ◽  
Fitting Model

Abstract Background Vietnam is undergoing an unprecedented pace of aging process and is expected to experience the fastest aging process in region. Association between increasing age and health deterioration has been well-documented across settings. Consequently, demand for healthcare utilization is rising among older people. However, healthcare utilization, here measured as count data, creates challenges for modeling because such data typically has distributions that are skewed with a large mass at zero. This study compares empirical econometric strategies for the modeling of healthcare utilization (measured as the number of outpatient visits in the last 12 months), and identifies the determinants of healthcare utilization among Vietnamese older people based on the best-fitting model identified. Methods Using the Vietnam Household Living Standard Survey in 2006 (N = 2426), nine econometric regression models for count data were examined to identify the best-fitting one. We used model selection criteria; statistical tests; and goodness-of-fit for in-sample model selection. In addition, we conducted 10-fold cross-validation checks to examine reliability of in-sample model selection. Finally, we utilized marginal effects to identify the factors associated with number of outpatient visits among Vietnamese older people based on the best-fitting model identified. Results We found strong evidence in favor of hurdle negative binomial model 2 (HNB2) for both in-sample selection and 10-fold cross-validation checks. The marginal effect results of the HNB2 showed that predisposing, enabling, need, and lifestyle factors were significantly associated with number of outpatient visits. The predicted probabilities for each count event showed the distinct trends of healthcare utilization among specific groups: at low count events, women and people in younger age group used more healthcare utilization than did men and their counterparts in older age groups, but a reversed trend was found at higher count events. Conclusions The findings here suggest that the HNB2 model should be considered for use in modeling counts of healthcare use. This study’s findings lay the groundwork for future research on the modeling of healthcare utilization in developing countries and those findings could be used to forecast on healthcare demand and making provisions for healthcare costs.

scholarly journals Assessing Model Selection Uncertainty Using a Bootstrap Approach: An Update

2016 ◽  
Vol 24 (2) ◽  
pp. 230-245 ◽  
Author(s):  
Gitta H. Lubke ◽  
Ian Campbell ◽  
Dan McArtor ◽  
Patrick Miller ◽  
Justin Luningham ◽  
...  
Keyword(s):  
Model Selection ◽  
Bootstrap Approach ◽  
Model Selection Uncertainty

scholarly journals Modeling count data for health care utilization: an empirical study of outpatient visits among Vietnamese older people

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Duc Dung Le ◽  
Roberto Leon Gonzalez ◽  
Joseph Upile Matola
Keyword(s):  
Model Selection ◽  
Older People ◽  
Healthcare Utilization ◽  
Count Data ◽  
Cross Validation ◽  
Negative Binomial ◽  
Outpatient Visits ◽  
Best Fitting ◽  
Fold Cross Validation ◽  
Fitting Model

Abstract Background Vietnam is undergoing a fast-aging process that poses potential critical issues for older people and central among those is demand for healthcare utilization. However, healthcare utilization, here measured as count data, creates challenges for modeling because such data typically has distributions that are skewed with a large mass at zero. This study compares empirical econometric strategies for the modeling of healthcare utilization (measured as the number of outpatient visits in the last 12 months) and identifies the determinants of healthcare utilization among Vietnamese older people based on the best-fitting model identified. Methods Using the Vietnam Household Living Standard Survey in 2006 (N = 2426), nine econometric regression models for count data were examined to identify the best-fitting one. We used model selection criteria, statistical tests and goodness-of-fit for in-sample model selection. In addition, we conducted 10-fold cross-validation checks to examine reliability of the in-sample model selection. Finally, we utilized marginal effects to identify the factors associated with the number of outpatient visits among Vietnamese older people based on the best-fitting model identified. Results We found strong evidence in favor of hurdle negative binomial model 2 (HNB2) for both in-sample selection and 10-fold cross-validation checks. The marginal effect results of the HNB2 showed that ethnicity, region, household size, health insurance, smoking status, non-communicable diseases, and disability were significantly associated with the number of outpatient visits. The predicted probabilities for each count event revealed the distinct trends of healthcare utilization among specific groups: at low count events, women and people in the younger age group used more healthcare utilization than did men and their counterparts in older age groups, but a reverse trend was found at higher count events. Conclusions The high degree of skewness and dispersion that typically characterizes healthcare utilization data affects the appropriateness of the econometric models that should be used in modeling such data. In the case of Vietnamese older people, our study findings suggest that hurdle negative binomial models should be used in the modeling of healthcare utilization given that the data-generating process reflects two different decision-making processes.

scholarly journals Modeling Count Data for Healthcare Utilization: an Empirical Study of Outpatient Visits Among Vietnamese Older People

2020 ◽  
Author(s):  
Dung Duc Le ◽  
Roberto Leon-Gonzalez ◽  
Joseph Upile Matola
Keyword(s):  
Model Selection ◽  
Older People ◽  
Healthcare Utilization ◽  
Count Data ◽  
Selection Criteria ◽  
Cross Validation ◽  
Outpatient Visits ◽  
Best Fitting ◽  
Fold Cross Validation ◽  
Fitting Model

Abstract Background Vietnam is undergoing a fast aging process that poses potential critical issues for older people and central among those is demand for healthcare utilization. However, healthcare utilization, here measured as count data, creates challenges for modeling because such data typically has distributions that are skewed with a large mass at zero. This study compares empirical econometric strategies for the modeling of healthcare utilization (measured as the number of outpatient visits in the last 12 months), and identifies the determinants of healthcare utilization among Vietnamese older people based on the best-fitting model identified. Methods Using the Vietnam Household Living Standard Survey in 2006 (N=2426), nine econometric regression models for count data were examined to identify the best-fitting one. We used model selection criteria; statistical tests; and goodness-of-fit for in-sample model selection. In addition, we conducted 10-fold cross-validation checks to examine reliability of in-sample model selection. Finally, we utilized marginal effects to identify the factors associated with the number of outpatient visits among Vietnamese older people based on the best-fitting model identified. Results We found strong evidence in favor of hurdle negative binomial model 2 (HNB2) for both in-sample selection and 10-fold cross-validation checks. The marginal effect results of the HNB2 showed that ethnicity, region, household size, health insurance, smoking status, non-communicable diseases, and disability were significantly associated with the number of outpatient visits. The predicted probabilities for each count event showed the distinct trends of healthcare utilization among specific groups: at low count events, women and people in the younger age group used more healthcare utilization than did men and their counterparts in older age groups, but a reversed trend was found at higher count events. Conclusions Data come in all shapes and sizes, this study highlights the importance of model specification checks and model selection criteria to avoid potential biased estimates as a result of model misspecifications. This study’s findings lay the groundwork for future research on the modeling of healthcare utilization in developing countries and those findings could be used to forecast on healthcare demand and making provisions for healthcare costs.

165 Effects of Digestible Lysine Level on Growth Performance and Economics of Grow-finish Pigs

2021 ◽  
Vol 99 (Supplement_1) ◽  
pp. 58-59
Author(s):  
Larissa L Becker ◽  
Emily E Scholtz ◽  
Joel M DeRouchey ◽  
Mike D Tokach ◽  
Jason C Woodworth ◽  
...  
Keyword(s):  
Phase 1 ◽  
Information Criterion ◽  
Response Curves ◽  
Phase 3 ◽  
Ileal Digestibility ◽  
Distillers Dried Grains ◽  
Feed Cost ◽  
Best Fitting ◽  
Dietary Treatments ◽  
Fitting Model

Abstract A total of 2,124 barrows and gilts (PIC 1050′DNA 600, initially 48.9 kg) were used in a 32-d study to determine the optimal dietary standardized ileal digestibility (SID) Lys level in a commercial setting. Pigs were randomly allotted to 1 of 5 dietary treatments with 24 to 27 pigs/pen and 16 replications/treatment. Similar number of barrows and gilts were placed in each pen. Diets were fed over 3 phases (48.9 to 58.6, 58.6 to 70.9, and 70.9 to 80.8 kg respectively). Dietary treatments were corn-soybean meal-based and contained 10 (phase 1 and 2) or 5% (phase 3) distillers dried grains with solubles. Diets were formulated to 85, 95, 103, 110, or 120% of the current Pig Improvement Company (PIC, Hendersonville, TN) SID Lys gilt recommendations with phase 1 SID Lys levels of 0.90, 1.01, 1.09, 1.17 and 1.27%, phase 2 levels of 0.79, 0.87, 0.94, 1.03, and 1.10%, and phase 3 levels of 0.71, 0.78, 0.85, 0.92, and 0.99%, respectively. Dose response curves were evaluated using linear (LM), quadratic polynomial (QP), broken-line linear (BLL), and broken-line quadratic (BLQ) models. For each response variable, the best-fitting model was selected using the Bayesian information criterion. Overall (d 0 to 32), increasing SID Lys increased (linear, P< 0.001) BW, ADG, G:F, Lys intake/d, and Lys intake/kg of gain. Modeling margin over feed cost (MOFC), BLL and QP estimated the requirement at 105.8% and 113.7% respectively. In summary, while growth increased linearly up to 120% of the PIC current feeding level, the optimal MOFC was 106% to 114% depending on the model used.

Genetic and Environmental Influences of Dietary Indices in a UK Female Twin Cohort

2020 ◽  
Vol 23 (6) ◽  
pp. 330-337
Author(s):  
Olatz Mompeo ◽  
Rachel Gibson ◽  
Paraskevi Christofidou ◽  
Tim D. Spector ◽  
Cristina Menni ◽  
...  
Keyword(s):  
Healthy Eating ◽  
Menopausal Status ◽  
Healthy Eating Index ◽  
Healthy Diet ◽  
Nutritional Epidemiology ◽  
Best Fitting ◽  
Heritability Estimates ◽  
Fitting Model ◽  
Dietary Indices ◽  
Health Parameters

AbstractA healthy diet is associated with the improvement or maintenance of health parameters, and several indices have been proposed to assess diet quality comprehensively. Twin studies have found that some specific foods, nutrients and food patterns have a heritable component; however, the heritability of overall dietary intake has not yet been estimated. Here, we compute heritability estimates of the nine most common dietary indices utilized in nutritional epidemiology. We analyzed 2590 female twins from TwinsUK (653 monozygotic [MZ] and 642 dizygotic [DZ] pairs) who completed a 131-item food frequency questionnaire (FFQ). Heritability estimates were computed using structural equation models (SEM) adjusting for body mass index (BMI), smoking status, Index of Multiple Deprivation (IMD), physical activity, menopausal status, energy and alcohol intake. The AE model was the best-fitting model for most of the analyzed dietary scores (seven out of nine), with heritability estimates ranging from 10.1% (95% CI [.02, .18]) for the Dietary Reference Values (DRV) to 42.7% (95% CI [.36, .49]) for the Alternative Healthy Eating Index (A-HEI). The ACE model was the best-fitting model for the Healthy Diet Indicator (HDI) and Healthy Eating Index 2010 (HEI-2010) with heritability estimates of 5.4% (95% CI [−.17, .28]) and 25.4% (95% CI [.05, .46]), respectively. Here, we find that all analyzed dietary indices have a heritable component, suggesting that there is a genetic predisposition regulating what you eat. Future studies should explore genes underlying dietary indices to further understand the genetic disposition toward diet-related health parameters.

Comparing productive vocabulary measures from the CDI and a systematic diary study

1999 ◽  
Vol 26 (1) ◽  
pp. 177-185 ◽  
Author(s):  
BYRON F. ROBINSON ◽  
CAROLYN B. MERVIS
Keyword(s):  
Growth Curves ◽  
Diary Study ◽  
Expressive Vocabulary ◽  
Early Language ◽  
Vocabulary Size ◽  
Early Language Development ◽  
Closed Class ◽  
Communicative Development ◽  
Best Fitting ◽  
Fitting Model

Expressive vocabulary data gathered during a systematic diary study of one male child's early language development are compared to data that would have resulted from longitudinal administration of the MacArthur Communicative Development Inventories spoken vocabulary checklist (CDI). Comparisons are made for (1) the number of words at monthly intervals (9;10.15 to 2;0.15), (2) proportion of words by lexical class (i.e. noun, predicate, closed class, ‘other’), (3) growth curves. The CDI underestimates the number of words in the diary study, with the underestimation increasing as vocabulary size increases. The proportion of diary study words appearing on the CDI differed as a function of lexical class. Finally, despite the differences in vocabulary size, logistic curves proved to be the best fitting model to characterize vocabulary development as measured by both the diary study and the CDI. Implications for the longitudinal use of the CDI are discussed.

scholarly journals Identification-Robust Estimates of the Canadian Natural Rate of Interest

2016 ◽  
Vol 91 (1-2) ◽  
pp. 161-176
Author(s):  
Maral Kichian
Keyword(s):  
Interest Rate ◽  
Robust Methods ◽  
Errors In Variables ◽  
Natural Rate ◽  
Real Rate ◽  
Point Estimates ◽  
Natural Rate Of Interest ◽  
Best Fitting ◽  
Estimate Precision ◽  
Fitting Model

The natural rate of interest is an unobservable entity and its measurement presents some important empirical challenges. In this paper, we use identification-robust methods and central bank real-time staff projections to obtain estimates for the equilibrium real rate from contemporaneous and forward-looking Taylor-type interest rate rules. The methods notably account for the potential presence of endogeneity, under-identification, and errors-in-variables concerns. Our applications are conducted on Canadian data. The results reveal some important identification difficulties associated with some of our models, reinforcing the need to use identification-robust methods to estimate such policy functions. Despite these challenges, we are able to obtain fairly comparable point estimates for the real equilibrium interest rate across our different models, and in the case of the best fitting model, also remarkable estimate precision.

Sign in / Sign up

Export Citation Format

Share Document