Rodrigo S. Targino
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Rodrigo S. Targino has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Gareth William Peters
University of California Santa Barbara
5 shared publications - Pavel V. Shevchenko
Department of Actuarial Studies and Business Analytics, Macquarie University, Sydney, Australia
5 shared publications Yuri Saporito
Getulio Vargas Foundation (FGV) - EMAp - School of Applied Mathematics
5 shared publications- Silvana M. Pesenti
Department of Statistical Sciences University of Toronto Canada
4 shared publications - Eduardo Fraga Lima de Melo
School of Applied Mathematics
3 shared publications - Luis Enrique Nieto Barajas
Department of Statistics ITAM Mexico
3 shared publications
Latest Publications
- Open Access
- 3 November 2026
- Published by Elsevier in SSRN Electronic Journal
Risk-Budgeted Mean-Variance Portfolios
Mean-variance portfolios exploit expected returns but concentrate risk, while risk budgeting controls how risk is shared but ignores expected returns. We propose the Risk-Budgeted Mean-Variance (RBMV) portfolio, which adds a minimum expected return and a maximum volatility to the risk budgeting problem. The resulting program is convex, has a unique solution when feasible, and nests risk budgeting. The portfolio's distance to the risk budget traces a cost curve showing how much risk budgeting is given up for each extra unit of expected return. Out of sample, in U.S. stocks, industry portfolios, and multi-asset ETFs, RBMV earns higher average returns than risk parity, with much lower turnover and concentration than mean-variance portfolios.
- 3 November 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessRisk-Budgeted Mean-Variance Portfolios
Mean-variance portfolios exploit expected returns but concentrate risk, while risk budgeting controls how risk is shared but ignores expected returns. We propose the Risk-Budgeted Mean-Variance (RBMV) portfolio, which adds a minimum expected return and a maximum volatility to the risk budgeting problem. The resulting program is convex, has a unique solution when feasible, and nests risk budgeting. The portfolio's distance to the risk budget traces a cost curve showing how much risk budgeting is given up for each extra unit of expected return. Out of sample, in U.S. stocks, industry portfolios, and multi-asset ETFs, RBMV earns higher average returns than risk parity, with much lower turnover and concentration than mean-variance portfolios.
- 10 May 2026
- Published by Escola de Negócios e Seguros in Revista Brasileira de Risco e Seguro
Uma aplicação prática de Precificação de Letra de Risco de Seguro (LRS) de Longevidade
A Letra de Risco de Seguro (LRS) é um instrumento recentemente regulamentado no Brasil. Equivalente aos chamados ILS – Insurance Linked Securities, ela se destina a ser mais uma alternativa de transferência de riscos, incluindo os de previdência. Há, no país, um grande interesse em fomentar conversão em renda (aposentadoria) nos planos de previdência complementar. Essa conversão costuma trazer risco de longevidade às entidades de previdência e às seguradoras. Neste sentido, o uso da LRS de Longevidade pode ajudar na gestão desse risco, auxiliando para que se atinja o objetivo de fomento. Neste artigo, precificamos LRS de Longevidade de um plano de benefícios com uma carteira real de aposentados. O título estruturado neste artigo cobre tanto o risco sistemático quanto o risco diversificável. Projetamos a evolução da população com o modelo Lee-Carter (1992) na abordagem Bayesiana. Computamos o prêmio para a cedente e a Taxa Interna de Retorno – TIR máxima para os investidores em uma variedade de cenários com diferentes gatilhos e maturidades da LRS. A exigência de que o limite máximo de pagamento à cedente deva ser completamente financiado na emissão da Letra torna importante o estabelecimento de um teto (cap) para que a TIR máxima seja de maior interesse para investidores.
- 10 May 2026
- Published by Escola de Negócios e Seguros in Revista Brasileira de Risco e Seguro
Uma aplicação prática de Precificação de Letra de Risco de Seguro (LRS) de Longevidade
A Letra de Risco de Seguro (LRS) é um instrumento recentemente regulamentado no Brasil. Equivalente aos chamados ILS – Insurance Linked Securities, ela se destina a ser mais uma alternativa de transferência de riscos, incluindo os de previdência. Há, no país, um grande interesse em fomentar conversão em renda (aposentadoria) nos planos de previdência complementar. Essa conversão costuma trazer risco de longevidade às entidades de previdência e às seguradoras. Neste sentido, o uso da LRS de Longevidade pode ajudar na gestão desse risco, auxiliando para que se atinja o objetivo de fomento. Neste artigo, precificamos LRS de Longevidade de um plano de benefícios com uma carteira real de aposentados. O título estruturado neste artigo cobre tanto o risco sistemático quanto o risco diversificável. Projetamos a evolução da população com o modelo Lee-Carter (1992) na abordagem Bayesiana. Computamos o prêmio para a cedente e a Taxa Interna de Retorno – TIR máxima para os investidores em uma variedade de cenários com diferentes gatilhos e maturidades da LRS. A exigência de que o limite máximo de pagamento à cedente deva ser completamente financiado na emissão da Letra torna importante o estabelecimento de um teto (cap) para que a TIR máxima seja de maior interesse para investidores.
- Open Access
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
Annuity Table Validation in Pension Funds: A Comparative Power Analysis of Goodness-of-Fit Tests
Assessing the goodness-of-fit of an annuity table to the observed mortality in a pension fund is a fundamental step in actuarial management, with direct impact on the actuarial valuation of benefits, provisioning, and risk model validation. In this paper, we compare the efficiency of different hypothesis tests, both those commonly used by the industry-such as the chi-squared test and the so-called Kolmogorov-Smirnov (KS) test-and the alternatives we propose in this article. Our proposals assume that death counts by age are modelled through a Poisson generalized linear model with offset and a deflator parameter that is either constant or linear in age, under both frequentist and Bayesian approaches. A comprehensive simulation study was carried out in three modalities: (i) simulation with age-constant deflators; (ii) simulation with age-linear deflators; and (iii) direct simulation from pairs of real annuity tables, in which the data-generating process corresponds to a known table and the null hypothesis is evaluated against a second also known table. The results show that some of our proposals exhibit type I error rates close to the nominal level while minimising type II error compared to the tests currently in common use in the industry. The results indicate that, unlike the tests we propose, the traditional chi-squared and KS tests inflate type I error and exhibit low power (high type II error), especially for populations that are not large and when the annuity table has higher mortality than the fund's experience-precisely the scenario requiring the most care from a solvency standpoint for survival benefit plans. To the best of our knowledge, the tests we propose in this article are novel and have never been subject of prior research.
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAnnuity Table Validation in Pension Funds: A Comparative Power Analysis of Goodness-of-Fit Tests
Assessing the goodness-of-fit of an annuity table to the observed mortality in a pension fund is a fundamental step in actuarial management, with direct impact on the actuarial valuation of benefits, provisioning, and risk model validation. In this paper, we compare the efficiency of different hypothesis tests, both those commonly used by the industry-such as the chi-squared test and the so-called Kolmogorov-Smirnov (KS) test-and the alternatives we propose in this article. Our proposals assume that death counts by age are modelled through a Poisson generalized linear model with offset and a deflator parameter that is either constant or linear in age, under both frequentist and Bayesian approaches. A comprehensive simulation study was carried out in three modalities: (i) simulation with age-constant deflators; (ii) simulation with age-linear deflators; and (iii) direct simulation from pairs of real annuity tables, in which the data-generating process corresponds to a known table and the null hypothesis is evaluated against a second also known table. The results show that some of our proposals exhibit type I error rates close to the nominal level while minimising type II error compared to the tests currently in common use in the industry. The results indicate that, unlike the tests we propose, the traditional chi-squared and KS tests inflate type I error and exhibit low power (high type II error), especially for populations that are not large and when the annuity table has higher mortality than the fund's experience-precisely the scenario requiring the most care from a solvency standpoint for survival benefit plans. To the best of our knowledge, the tests we propose in this article are novel and have never been subject of prior research.
- Open Access
Negative binomial models for development triangles of counts
Prediction of outstanding claims has been done via nonparametric models (chain ladder), semiparametric models (overdispersed poisson) or fully parametric models. In this paper, we propose models based on negative binomial distributions for the prediction of outstanding number of claims, which are particularly useful to account for overdispersion. We first assume independence of random variables and introduce appropriate notation. Later, we generalise the model to account for dependence across development years. In both cases, the marginal distributions are negative binomials. We study the properties of the models and carry out bayesian inference. We illustrate the performance of the models with simulated and real datasets.
Open AccessNegative binomial models for development triangles of counts
Prediction of outstanding claims has been done via nonparametric models (chain ladder), semiparametric models (overdispersed poisson) or fully parametric models. In this paper, we propose models based on negative binomial distributions for the prediction of outstanding number of claims, which are particularly useful to account for overdispersion. We first assume independence of random variables and introduce appropriate notation. Later, we generalise the model to account for dependence across development years. In both cases, the marginal distributions are negative binomials. We study the properties of the models and carry out bayesian inference. We illustrate the performance of the models with simulated and real datasets.
- 29 November 2025
- Published by Elsevier in International Journal of Forecasting
- Vol. 42 (2) , 691-707
Stochastic modelling of football matches using dynamic regressors
- 29 November 2025
- Published by Elsevier in International Journal of Forecasting
- Vol. 42 (2) , 691-707
Stochastic modelling of football matches using dynamic regressors
- 27 October 2025
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 53 (9) , 1628-1647
Conformal prediction for frequency-severity modeling
- 27 October 2025
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 53 (9) , 1628-1647
Conformal prediction for frequency-severity modeling
- Open Access
- 17 May 2025
- Published by Fundacao Getulio Vargas in Brazilian Review of Finance
- Vol. 23, e202508
Mensuração da transferência de riqueza em planos de contribuição definida com a marcação de ativos na curva
O método de mensuração de ativos financeiros em planos de previdência na modalidade de contribuição definida (CD, ou contribuição variável - CV, na fase de acumulação) possui implicações significativas sobre eventuais transferências de riqueza entre os participantes. Em Dez/2024 foi publicada norma para as Entidades Fechadas de Previdência Complementar possibilitando o uso da marcação na curva de títulos públicos federais nos planos CD e CV na fase de acumulação. Este artigo quantifica o impacto financeiro nos participantes da adoção da marcação na curva (HTM – Hold to Maturity) nestes planos, utilizando dados reais da estrutura a termo da taxa de juros de cupom de IPCA para avaliar as transferências de riqueza resultantes dessa adoção. A análise evidencia como a marcação na curva gera assimetrias nos resultados financeiros, beneficiando alguns participantes em detrimento de outros. As transferências de riqueza ocorrem tanto em qualquer retirada de recursos quanto também na entrada (contribuições), inclusive realocações da carteira que impliquem venda ou compra de títulos. O uso do HTM de forma parcial ou a tentativa de imunização de saídas não eliminam por completo transferências de riqueza. Os resultados reforçam que, para fins de cotização, o uso da marcação a mercado (MTM - Mark to Market) de ativos em planos CD (e CV na fase de diferimento) evita transferências de riqueza e, por consequência, prejuízos financeiros aos seus participantes.
- 17 May 2025
- Published by Fundacao Getulio Vargas in Brazilian Review of Finance
- Vol. 23, e202508
Open AccessMensuração da transferência de riqueza em planos de contribuição definida com a marcação de ativos na curva
O método de mensuração de ativos financeiros em planos de previdência na modalidade de contribuição definida (CD, ou contribuição variável - CV, na fase de acumulação) possui implicações significativas sobre eventuais transferências de riqueza entre os participantes. Em Dez/2024 foi publicada norma para as Entidades Fechadas de Previdência Complementar possibilitando o uso da marcação na curva de títulos públicos federais nos planos CD e CV na fase de acumulação. Este artigo quantifica o impacto financeiro nos participantes da adoção da marcação na curva (HTM – Hold to Maturity) nestes planos, utilizando dados reais da estrutura a termo da taxa de juros de cupom de IPCA para avaliar as transferências de riqueza resultantes dessa adoção. A análise evidencia como a marcação na curva gera assimetrias nos resultados financeiros, beneficiando alguns participantes em detrimento de outros. As transferências de riqueza ocorrem tanto em qualquer retirada de recursos quanto também na entrada (contribuições), inclusive realocações da carteira que impliquem venda ou compra de títulos. O uso do HTM de forma parcial ou a tentativa de imunização de saídas não eliminam por completo transferências de riqueza. Os resultados reforçam que, para fins de cotização, o uso da marcação a mercado (MTM - Mark to Market) de ativos em planos CD (e CV na fase de diferimento) evita transferências de riqueza e, por consequência, prejuízos financeiros aos seus participantes.
- Open Access
Reverse-BSDE Monte Carlo
Recently, there has been a growing interest in generative models based on diffusions driven by the empirical robustness of these methods in generating high-dimensional photorealistic images and the possibility of using the vast existing toolbox of stochastic differential equations. %This remarkable ability may stem from their capacity to model and generate multimodal distributions. In this work, we offer a novel perspective on the approach introduced in Song et al. (2021), shifting the focus from a "learning" problem to a "sampling" problem. To achieve this, we reformulate the equations governing diffusion-based generative models as a Forward-Backward Stochastic Differential Equation (FBSDE), which avoids the well-known issue of pre-estimating the gradient of the log target density. The solution of this FBSDE is proved to be unique using non-standard techniques. Additionally, we propose a numerical solution to this problem, leveraging on Deep Learning techniques. This reformulation opens new pathways for sampling multidimensional distributions with densities known up to a normalization constant, a problem frequently encountered in Bayesian statistics.
Open AccessReverse-BSDE Monte Carlo
Recently, there has been a growing interest in generative models based on diffusions driven by the empirical robustness of these methods in generating high-dimensional photorealistic images and the possibility of using the vast existing toolbox of stochastic differential equations. %This remarkable ability may stem from their capacity to model and generate multimodal distributions. In this work, we offer a novel perspective on the approach introduced in Song et al. (2021), shifting the focus from a "learning" problem to a "sampling" problem. To achieve this, we reformulate the equations governing diffusion-based generative models as a Forward-Backward Stochastic Differential Equation (FBSDE), which avoids the well-known issue of pre-estimating the gradient of the log target density. The solution of this FBSDE is proved to be unique using non-standard techniques. Additionally, we propose a numerical solution to this problem, leveraging on Deep Learning techniques. This reformulation opens new pathways for sampling multidimensional distributions with densities known up to a normalization constant, a problem frequently encountered in Bayesian statistics.
- 1 May 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (3) , 1208-1229
Risk Budgeting Allocation for Dynamic Risk Measures
Time-Consistent Diversified Portfolio Allocations Portfolio allocation problems typically involve diversification of risks, which can be achieved by risk budgeting strategies. Of additional importance is being time consistent—that is, ensuring that decisions about what to do in the future under certain outcomes remain optimal when those future outcomes are realized. In “Risk Budgeting Allocation for Dynamic Risk Measures,” Pesenti, Jaimungal, Saporito, and Targino develop time-consistent portfolio strategies, termed dynamic risk budgeting strategies, where the diversification is on the level of risk contribution of assets. The authors formalize the dynamic setting of risk budgeting strategies, recast dynamic risk budgeting strategies as solutions of sequences of convex optimization problems, and develop an efficient actor-critic deep reinforcement learning algorithm for estimating dynamic risk budgeting strategies. The algorithm and the dynamic risk budgeting strategies are illustrated on a complex simulation case study involving five assets and 12 time steps.
- 1 May 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (3) , 1208-1229
Risk Budgeting Allocation for Dynamic Risk Measures
Time-Consistent Diversified Portfolio Allocations Portfolio allocation problems typically involve diversification of risks, which can be achieved by risk budgeting strategies. Of additional importance is being time consistent—that is, ensuring that decisions about what to do in the future under certain outcomes remain optimal when those future outcomes are realized. In “Risk Budgeting Allocation for Dynamic Risk Measures,” Pesenti, Jaimungal, Saporito, and Targino develop time-consistent portfolio strategies, termed dynamic risk budgeting strategies, where the diversification is on the level of risk contribution of assets. The authors formalize the dynamic setting of risk budgeting strategies, recast dynamic risk budgeting strategies as solutions of sequences of convex optimization problems, and develop an efficient actor-critic deep reinforcement learning algorithm for estimating dynamic risk budgeting strategies. The algorithm and the dynamic risk budgeting strategies are illustrated on a complex simulation case study involving five assets and 12 time steps.
- 7 June 2023
- Published by Elsevier in European Journal of Operational Research
- Vol. 311 (3) , 1040-1056
Risk budgeting portfolios from simulations
- 7 June 2023
- Published by Elsevier in European Journal of Operational Research
- Vol. 311 (3) , 1040-1056
Risk budgeting portfolios from simulations