Dynamic regime marginal structural mean models for estimation of optimal dynamic treatment regimes, Part I: main content

Dynamic treatment regimes are set rules for sequential decision making based on patient covariate history. Observational studies are well suited for the investigation of the effects of dynamic treatment regimes because of the variability in treatment decisions found in them. This variability exists because different physicians make different decisions in the face of similar patient histories. In this article we describe an approach to estimate the optimal dynamic treatment regime among a set of enforceable regimes. This set is comprised by regimes defined by simple rules based on a subset of past information. The regimes in the set are indexed by a Euclidean vector. The optimal regime is the one that maximizes the expected counterfactual utility over all regimes in the set. We discuss assumptions under which it is possible to identify the optimal regime from observational longitudinal data. Murphy et al. (2001) developed efficient augmented inverse probability weighted estimators of the expected utility of one fixed regime. Our methods are based on an extension of the marginal structural mean model of Robins (1998, 1999) which incorporate the estimation ideas of Murphy et al. (2001). Our models, which we call dynamic regime marginal structural mean models, are specially suitable for estimating the optimal treatment regime in a moderately small class of enforceable regimes of interest. We consider both parametric and semiparametric dynamic regime marginal structural models. We discuss locally efficient, double-robust estimation of the model parameters and of the index of the optimal treatment regime in the set. In a companion paper in this issue of the journal we provide proofs of the main results.

Dynamic regime marginal structural mean models for estimation of optimal dynamic treatment regimes, Part II: proofs of results

In this companion article to "Dynamic Regime Marginal Structural Mean Models for Estimation of Optimal Dynamic Treatment Regimes, Part I: Main Content" [Orellana, Rotnitzky and Robins (2010), IJB, Vol. 6, Iss. 2, Art. 7] we present (i) proofs of the claims in that paper, (ii) a proposal for the computation of a confidence set for the optimal index when this lies in a finite set, and (iii) an example to aid the interpretation of the positivity assumption.

Marginal and Interaction Effects in Ordered Response Models

In discrete choice models the marginal effect of a variable of interest that is interacted with another variable differs from the marginal effect of a variable that is not interacted with any variable. The magnitude of the interaction effect is also not equal to the marginal effect of the interaction term. I present consistent estimators of both marginal and interaction effects in ordered response models. This procedure is general and can easily be extended to other discrete choice models.

Marginal efficiencies of long chain-polyunsaturated fatty acid use by barramundi (Lates calcarifer) when fed diets with varying blends of fish oil and poultry fat

An experiment was conducted with barramundi (Lates calcarifer) juveniles to examine the marginal efficiency of utilisation of long chain-polyunsaturated fatty acids (LC-PUFA). A series of five diets with blends of fish (anchovy) oil and poultry fat (F100:P0, F60:P40, F30:P70, F15:P85, F0:P100) were fed to 208. ±. 4.1. g fish over a 12-week period. The replacement of fish oil with poultry fat had no impact on growth performance (average final weight of 548.3. ±. 10.2. g) or feed conversion (mean = 1.14. ±. 0.02). Analysis of the whole body composition showed that the fatty acid profile reflected that of the fed diet. However it was also shown that there was a disproportional retention of some fatty acids relative to others (notably LOA, 18:2n-6 and LNA, 18:3n-3). By examining the body mass independent retention of different fatty acids with differential levels of intake of each, the marginal efficiencies of the use of these nutrients by this species were able to be determined. The differential retention of fatty acids in the meat was also examined allowing the determination of oil blending strategies to optimise meat n-3 LC-PUFA levels.

Estimação de regressões aditivas via backfitting e integração marginal: performance em pequenas amostras

Nesta dissertação realizou-se um experimento de Monte Carlo para re- velar algumas características das distribuições em amostras finitas dos estimadores Backfitting (B) e de Integração Marginal(MI) para uma regressão aditiva bivariada. Está-se particularmente interessado em fornecer alguma evidência de como os diferentes métodos de seleção da janela hn, tais co- mo os métodos plug-in, impactam as propriedades em pequenas amostras dos estimadores. Está-se interessado, também, em fornecer evidência do comportamento de diferentes estimadores de hn relativamente a seqüência ótima de hn que minimiza uma função perda escolhida. O impacto de ignorar a dependência entre os regressores na estimação da janela é tam- bém investigado. Esta é uma prática comum e deve ter impacto sobre o desempenho dos estimadores. Além disso, não há nenhuma rotina atual- mente disponível nos pacotes estatísticos/econométricos para a estimação de regressões aditivas via os métodos de Backfitting e Integração Marginal. É um dos objetivos a criação de rotinas em Gauss para a implementação prática destes estimadores. Por fim, diferentemente do que ocorre atual- mente, quando a utilização dos estimadores-B e MI é feita de maneira completamente ad-hoc, há o objetivo de fornecer a usuários informação que permita uma escolha mais objetiva de qual estimador usar quando se está trabalhando com uma amostra finita.

A estrutura subjacente ao fenômeno da produtividade marginal percebida em empresas de serviço

Esta pesquisa de dissertação teve como objetivo verificar a estrutura subjacente ao fenômeno da Produtividade Marginal Percebida. O estudo limitou-se ao fenômeno ocorrido em empresas de serviço, que não dispusessem de indicadores quantitativos de produtividade. O arcabouço teórico está fundamentado na literatura disponível sobre Produtividade e Comportamento Organizacional. O método utilizado foi o de pesquisa qualitativa de caráter estruturalista. O objetivo foi investigar a existência de estruturas subjacentes ao fenômeno da Produtividade Marginal Percebida. A coleta das informações ocorreu através de entrevistas semi-estruturadas com chefes de empresas de serviços no município do Rio de Janeiro. Os sujeitos foram definidos através de seleção acidental e limitados pelo critério de saturação. A análise das informações buscou permitir o desenvolvimento de um modelo que explicitasse os elementos constituintes dos esforços para elevar a produtividade do trabalho e suas relações, de forma a poder desvendar a estrutura subjacente ao sistema relacional da produtividade. O caráter desta pesquisa é exploratório, portanto não coloca de antemão hipóteses a serem testadas. Ao final da pesquisa, a interpretação das informações coletados sugeriu que a relação subjacente ao fenômeno aqui objeto de estudo uma relação de Influência Mútua, qualificada como não-sequencial, pois não há ordenação entre os elementos da relação, e segue a lógica da reflexividade. Esta lógica afirma que o fato de observar um fenômeno é suficiente para alterá-lo. O trabalho é concluído com a apresentação de críticas, limitações da pesquisa e sugestão para próximos trabalhos.

O efeito preço das transferências sobre o custo marginal de financiamento público do IPTU para os governos locais no Brasil

O objetivo desse artigo é estimar o efeito-preço das transferências incondicionais, condicionais e da renda para os municípios brasileiros. De acordo com Dahlby (2011) um aumento de transferência lump-sum tem, além do efeito renda, um efeito preço decorrente do uso de impostos distorcivos. Dessa forma, um governo local que recebe uma transferência lump-sum, pode diminuir o custo marginal de financiamento público (MCF) e permanecer com o mesmo nível de serviço. Assim, o efeito gasto das transferências pode ser maior do que o decorrente da renda, explicando o flypaper effect. Usando dados de impostos sobre propriedade (IPTU), primeiramente calculamos o custo marginal de financiamento público (MCF) deste imposto. Em seguida, estimamos se as transferências lump-sum efetivamente diminuem o custo marginal de financiamento público (MCF).

The distribution of average and marginal effective tax rates in European Union member states

Macro-based summary indicators of effective tax burdens do not capture differences in effective tax rates facing different sub-groups of the population. They also cannot provide information on the level or distribution of the marginal effective tax rates thought to influence household behaviour. I use EUROMOD, an EU-wide tax-benefit microsimulation model, to compute distributions of average and marginal effective tax rates across the household population in fourteen European Union Member States. Using different definitions of ‘net taxes’, the tax base and the unit of analysis I present a range of measures showing the contribution of the tax-benefit system to household incomes, the average effective tax rates applicable to income from labour and marginal effective tax rates faced by working men and women. In a second step, effective tax rates are broken down to separately show the influence of each type of tax-benefit instrument. The results show that measures of effective tax rates vary considerably depending on incomes, labour market situations and family circumstances. Using single averages or macro-based indicators will therefore provide an inappropriate picture of tax burdens faced by large parts of the population.

Three applications of the marginal cost of public funds concept

This paper illustrates the use of the marginal cost of public funds concept in three contexts. First, we extend Parry’s (2003) analysis of the efficiency effects excise taxes in the U.K., primarily by incorporating the distortion caused by imperfect competition in the cigarette market and distinguishing between the MCFs for per unit and ad valorem taxes on cigarettes. Our computations show, contrary to the standard result in the literature, that the per unit tax on cigarettes has a slightly lower MCF than the ad valorem tax on cigarettes. Second, we calculate the MCF for a payroll tax in a labour market with involuntary unemployment, using the Shapiro and Stiglitz (1984) efficiency wage model as our framework. Our computations, based on Canadian labour market data, indicate that incorporating the distortion caused by involuntary unemployment raises the MCF by 25 to 50 percent. Third, we derive expressions for the distributionally-weighted MCFs for the exemption level and the marginal tax rate for a “flat tax”, such as the one that has been adopted by the province of Alberta. This allows us to develop a restricted, but tractable, version of the optimal income tax problem. Computations indicate that the optimal marginal tax rate may be quite high, even with relatively modest pro-poor distributional preferences.

The marginal cost of funds from public sector borrowing

An expression for the welfare cost of a marginal increase in the public debt is derived using a simple AK endogenous growth model. This measure of the marginal cost of public funds (MCF) can be interpreted as the marginal benefit-cost ratio that a debtfinanced public project needs in order to generate a net social gain. The model predicts an increase in the public debt ratio will have little effect on the optimal public expenditure ratio and that most of the adjustment will occur on the tax side of the budget.

Bank relationships: effect on the availability and marginal cost of credit for firms in Argentina

The paper provides evidence on what affects at the margin the cost and availability of bank credit for firms in Argentina. We study in particular how banks use different pieces of private and public information to screen firms and overcome informational asymmetries in the credit market. Some private information is transferable, like balance sheet data. Private information generated in relationships is not. To capture the closeness of bank relationships, we resort to the concentration of bank credit and the number of credit lines in a bank. We also consider public information available in the Central de Deudores. The cost of credit is measured using overdrafts, the most expensive line of credit, at the bank that charges the highest rate for overdrafts. We find that the cost of credit is smaller for a firm with a close relationship to the marginal bank. Firms with large assets, a high sales/assets ratio, and a low debt/assets ratio pay a lower interest rate at the margin. A good credit history (no debt arrears and no bounced checks) and collateral also reduce the marginal interest rate. The availability of credit is measured by unused credit lines as a proportion of total liabilities with the main bank. The availability of credit depends positively on a close relationship with the main bank. Large assets, a high return over assets, a high sales/assets ratio, a low debt/assets ratio, a good credit history, and collateral lead to higher credit availability. Our measure of unused credit lines is less ambiguous than traditional measures like leverage, which may indicate financial distress rather than availability of credit.

Curva de Phillips novo keynesiana, custo marginal e expectativas de inflação no Brasil

A importância deste trabalho será analisar a dinâmica inflacionária brasileira de acordo com um modelo híbrido de curva de Phillips baseado nos estudos sobre a curva de Phillips “Novo-Keynesiana”. Através deste estudo será possível avaliar a inflação passada e as expectativas de inflação como determinantes à inflação corrente e a aplicabilidade da curva de Phillips “Novo-Keynesiana” para o caso brasileiro. As conclusões do estudo darão uma indicação mais concreta do novo dinamismo da inflação brasileira e do papel das expectativas de inflação como variável predissora.