| 1. |
- Andersson, Björn, 1984-, et al.
(author)
-
Sensitivity analysis of violations of the faithfulness assumption
- 2014
-
In: Journal of Statistical Computation and Simulation. - : Taylor & Francis Group. - 0094-9655 .- 1563-5163. ; 84:7, s. 1608-1620
-
Journal article (other academic/artistic)abstract
- We study implications of violations of the fatihfulness condition due to parameter cancellations on estimation of the DAG skeleton. Three settings are investigated: when i) faithfulness is guaranteed ii) faithfulness is not guaranteed and iii) the parameter distributions are concentrated around unfaithfulness (near-unfaithfulness). In a simulation study the effetcs of the different settings are compared using the PC and MMPC algorithms. The results show that the performance in the faithful case is almost unchanged compared to the unrestricted case whereas there is a general decrease in performance under the near-unfaithful case as compared to the unrestricted case. The response to near-unfaithful parameterisations is similar between two algorithms, with the MMPC algorithm having higher true positive rates and the PC algorithm having lower false positive rates.
|
|
| 2. |
|
|
| 3. |
- Hedström, Erik, et al.
(author)
-
Incidence of fractures among children and adolescents in rural and urban communities : analysis based on 9,965 fracture events
- 2014
-
In: Injury Epidemiology. - : Springer Science and Business Media LLC. - 2197-1714. ; 1
-
Journal article (peer-reviewed)abstract
- Background: Previous work has explored the significance of residence on injuries. A number of articles reported higher rates of injury in rural as compared to urban settings. This study aimed to evaluate the importance of residency on the occurrence of fractures among children and adolescents within a region in northern Sweden.Methods: In a population based study with data from an injury surveillance registry at a regional hospital, we have investigated the importance of sex, age and place of residency for the incidence of fractures among children and adolescents 0-19 years of age using a Poisson logistic regression analysis. Data was collected between 1998 and 2011.Results: The dataset included 9,965 cases. Children and adolescents growing up in the most rural communities appeared to sustain fewer fractures than their peers in an urban municipality, risk ratio 0.81 (0.76-0.86). Further comparisons of fracture rates in the urban and rural municipalities revealed that differences were most pronounced for sports related fractures and activities in school in the second decade of life.Conclusion: Results indicate that fracture incidence among children and adolescents is affected by place of residency. Differences were associated with activity at injury and therefore we have discussed the possibility that this effect was due to the influence of place on activity patterns.The results suggest it is of interest to explore how geographic and demographic variables affect the injury pattern further.
|
|
| 4. |
|
|
| 5. |
- Persson, Emma, 1981-, et al.
(author)
-
Estimating a marginal causal odds ratio in a case-control design : analyzing the effect of low birth weight on the risk of type 1 diabetes mellitus
- 2013
-
In: Statistics in Medicine. - : Wiley. - 0277-6715 .- 1097-0258. ; 32:14, s. 2500-2512
-
Journal article (peer-reviewed)abstract
- Estimation of marginal causal effects from case-control data has two complications: (i) confounding due to the fact that the exposure under study is not randomized, and (ii) bias from the case-control sampling scheme. In this paper, we study estimators of the marginal causal odds ratio, addressing these issues for matched and unmatched case-control designs when utilizing the knowledge of the known prevalence of being a case. The estimators are implemented in simulations where their finite sample properties are studied and approximations of their variances are derived with the delta method. Also, we illustrate the methods by analyzing the effect of low birth weight on the risk of type 1 diabetes mellitus using data from the Swedish Childhood Diabetes Register, a nationwide population-based incidence register.
|
|
| 6. |
- Waernbaum, Ingeborg, 1972-
(author)
-
Model misspecification and robustness in causal inference : comparing matching with doubly robust estimation
- 2012
-
In: Statistics in Medicine. - : Wiley. - 0277-6715 .- 1097-0258. ; 31:15, s. 1572-1581
-
Journal article (peer-reviewed)abstract
- In this paper we compare the robustness properties of a matching estimator with a doubly robust estimator. We describe the robustness properties of matching and subclassification estimators by showing how misspecification of the propensity score model cam result in consistent estimation of the average causal effect. The propensity scores are covariate scores, which are a class of functions that removes bias due to all observed covariates. When matching on a parametric model (e.g. a propensity or prognostic score), the matching estimator is robust to model misspecifications if the misspecified model belongs to the class of covariate scores. The implication is that there are multiple possibilities for the matching estimator in contrast to the doubly robust estimator in which the researcher has two chances to make reliable inference. In simulations, we compare the finite sample properties of the matching estimator with a simple inverse probability weighting estimator and a doubly robust estimator. For the misspecifications in our study the mean square error of the matching estimator is smaller than the mean square error of both the simple inverse probability weighting estimator and the doubly robust estimator-
|
|
| 7. |
- Waernbaum, Ingeborg, 1972-
(author)
-
Propensity score model specification for estimation of average treatment effects
- 2010
-
In: Journal of Statistical Planning and Inference. - Amsterdam : Elsevier. - 0378-3758 .- 1873-1171. ; 140:7, s. 1948-1956
-
Journal article (peer-reviewed)abstract
- Treatment effect estimators that utilize the propensity score as a balancing score, e.g., matching and blocking estimators are robust to misspecifications of the propensity score model when the misspecification is a balancing score. Such misspecifications arise from using the balancing property of the propensity score in the specification procedure. Here, we study misspecifications of a parametric propensity score model written as a linear predictor in a strictly monotonic function, e.g. a generalized linear model representation. Under mild assumptions we show that for misspecifications, such as not adding enough higher order terms or choosing the wrong link function, the true propensity score is a function of the misspecified model. Hence, the latter does not bring bias to the treatment effect estimator. It is also shown that a misspecification of the propensity score does not necessarily lead to less efficient estimation of the treatment effect. The results of the paper are highlighted in simulations where different misspecifications are studied.
|
|
