Vitor Augusto Ozaki

Spatio-Temporal Modeling of Agricultural Yield Data with an Application to Pricing Crop Insurance Contracts

This article presents a statistical model of agricultural yield data based on a set of hierarchical Bayesian models that allows joint modeling of temporal and spatial autocorrelation. This method captures a comprehensive range of the various uncertainties involved in predicting crop insurance premium rates as opposed to the more traditional ad hoc, two-stage methods that are typically based on independent estimation and prediction. A panel data set of county-average yield data was analyzed for 290 counties in the State of Paraná (Brazil) for the period of 1990 through 2002. Posterior predictive criteria are used to evaluate different model specifications. This article provides substantial improvements in the statistical and actuarial methods often applied to the calculation of insurance premium rates. These improvements are especially relevant to situations where data are limited.

Parametric and nonparametric statistical modelling of crop yield: implications for pricing crop insurance contracts

This article considers alternative methods to calculate the fair premium rate of crop insurance contracts based on county yields. The premium rate was calculated using parametric and nonparametric approaches to estimate the conditional agricultural yield density. These methods were applied to a data set of county yield provided by the Statistical and Geography Brazilian Institute (IBGE), for the period of 1990 through 2002, for soybean, corn and wheat, in the State of Paraná. In this article, we propose methodological alternatives to pricing crop insurance contracts resulting in more accurate premium rates in a situation of limited data.

Missing data imputation of climate datasets: implications to modeling extreme drought events

Time series from weather stations in Brazil have several missing data, outliers and spurious zeroes. In order to use this dataset in risk and meteorological studies, one should take into account alternative methodologies to deal with these problems. This article describes the statistical imputation and quality control procedures applied to a database of daily precipitation from meteorological stations located in the State of Parana, Brazil. After imputation, the data went through a process of quality control to identify possible errors, such as: identical precipitation over seven consecutive days and precipitation values that differ significantly from the values in neighboring weather stations. Next, we used the extreme value theory to model agricultural drought, considering the maximum number of consecutive days with precipitation below 7 mm for the period between January and February, in the main soybean agricultural regions in the State of Parana.

Em busca de um novo paradigma para o seguro rural no Brasil

The implementation of a crop insurance program is an old requirement of the agricultural sector. After the failure of the National Crop Insurance Company (CNSA), in the period of 1954 through 1966, the federal government started to stimulate the agricultural insurance market. Aiming increase the crop insurance operations the Government approved Law n. 10.823 which subsidize part of the premium paid by farmers. The nature of this work is essentially analytical, such that any quantitative model has been established. Over the years, government risk management tools, such as, Proagro and some private insurance companies have had unsatisfactory financial results suggesting that the current crop insurance model presents signals of decline. This article shows the main initiatives, state and private, and suggests a set of measures to develop the agricultural insurance in the country.

ANÁLISE ESPACIAL DA PRODUTIVIDADE AGRÍCOLA NO ESTADO DO PARANÁ: IMPLICAÇÕES PARA O SEGURO AGRÍCOLA

O presente estudo tem como objetivo analisar espacialmente os dados de produtividade agricola. Atraves da estatistica espacial e possivel estimar alguns parâmetros do semivariograma relevantes para estudar o problema do risco sistemico, com implicacoes nos programas de seguro agricola. Em particular estimou-se os parâmetros de alcance, para se verificar a distância, na qual a correlacao espacial tende a zero. Para a analise empirica, utilizou-se dados de produtividade agricola municipal do IBGE, para a soja e milho, no estado do Parana, no periodo de 1990 a 2002. O estudo mostrou que de fato a dependencia espacial destes dados existe, tornando-se praticamente nula a distâncias relativamente longas (em km) e pode ser captada em todos os anos, para ambas as culturas.

A distribuição normal-assimétrica como modelo para produtividade de milho aplicada ao seguro agrícola

The crop insurance sold in Brazil is estimated using the average of the municipal agricultural productivity of the last four or five years. Many researchers believe that using a probabilistic model for the distribution of agricultural productivity is a better methodology for this purpose. This study advocates using the skew-normal distribution as an alternative to the normal distribution to model the distribution of agricultural productivity in Brazil. We analyzed series of corn yield from 1981 to 2007, in 30 counties in Parana State, Brazil. The skew-normal distribution showed better fits than the normal distribution for the vast majority of the counties, and, consequently, led to better estimates for the expected payment of crop insurance.

Uma digressão sobre o programa de subvenção ao prêmio do seguro rural e as implicações para o futuro deste mercado

Nos ultimos anos, o governo federal tem incentivado o mercado de seguro rural por meio de medidas pontuais que visam o aumento da demanda e o estabelecimento de condicoes adequadas para as seguradoras ofertarem seus produtos. Uma dessas medidas foi o Programa de Subvencao ao Premio do Seguro Rural (PSP), que apresentou grande crescimento desde 2005, quando o programa comecou. Entretanto, este crescimento tem se concentrado apenas no estado do Parana. O artigo mostra que o entrave para expandir o seguro rural para outros estados nao se restringe apenas ao risco relativo, mas a inexistencia de uma base consolidada de dados nos municipios analisados.......Over the last years, the federal government has been stimulating the crop insurance market through a few measures that aim to increase the demand and support insurance companies to offer their products. One of these measures was the Insurance Premium Subsidy Program (PSP) that registered an increase since the beginning of the program. However, this expressive growth has been concentrating only in the Parana state. This article states that one of the main obstacles to expand the agricultural insurance to other states is not restricted only to the relative risk, but to the inexistence of a consolidated database in several cities.

Modeling of soybean yield using symmetric, asymmetric and bimodal distributions: implications for crop insurance

ABSTRACT Over the years, many papers used parametric distributions to model crop yields, such as: normal (N), Beta, Log-normal and the Skew-normal (SN). These models are well-defined, mathematically and also computationally, but its do not incorporate bimodality. Therefore, it is necessary to study distributions which are more flexible in modeling, since most of crop yield data in Brazil presents evidence of asymmetry or bimodality. Thus, the aim of this study was to model and forecast soybean yields for municipalities in the State of Paran, in the period from 1980 to 2014, using the Odd log normal logistic (OLLN) distribution for the bimodal data and the Beta, SN and Skew-t distributions for the symmetrical and asymmetrical series. The OLLN model was the one which best fit the data. The results were discussed in the context of crop insurance pricing.

Reduzindo a Incerteza no Mercado de Seguros: Uma Abordagem via Informações de Sensoriamento Remoto e Atuária

Em qualquer contrato de seguro, dois parâmetros sao fundamentais: a taxa de premio e a indenizacao. A metodologia de calculo da taxa de premio e fundamental para evitar problemas de assimetria de informacao, enquanto metodos de acompanhamento do objeto segurado podem ser uteis no dimensionamento e controle das perdas. Em geral, a incerteza do fluxo financeiro das empresas em um mercado contingente e elevada. O estudo propoe reduzir essa incerteza por meio de metodos alternativos de precificacao baseados em modelos hierarquicos Bayesianos e no uso de informacoes de sensoriamento remoto. A metodologia aprimora o entendimento da dinâmica temporal e espacial do fluxo financeiro de um agente economico no mercado de seguro agricola considerado um dos ramos mais complexos para se operacionalizar. Os resultados mostram que a metodologia de precificacao estima com relativa precisao as taxas de premio e o uso da geotecnologia aponta para melhoras significativas na quantificacao das perdas agricolas.

Estudo Sobre a Viabilidade do Seguro Receita: o caso da região oeste do Paraná

Since 2003 crop insurance programs have become the focus of agricultural policy in Brazil. Given the increasing interest in insurance, accurate calculation of premium values has great importance. We address the revenue distribution issue and its implication by pricing a new Brazilian insurance contract, considering the skew-normal distribution of probability of revenue data from the western region of Parana State (Brazil). Results show that insurers (empirical) rates tend to underestimate the risk by lower coverage levels (50% and 60%). At 70% of coverage level, the ER and SN rates seem the same on average. The estimated insured cash flows for 2006/2007 to 2010/2011crops, by coverage level, were positive for all seasons, indicating that agricultural operations with insurance can generate positive return for insurers, presenting as viable modality for insurance activity.