Mila Novita | University of Indonesia (original) (raw)
Papers by Mila Novita
Journal of physics, May 1, 2019
Family forms are many and varied, reflecting a myriad of understandings and influencing factors. ... more Family forms are many and varied, reflecting a myriad of understandings and influencing factors. In any given cultural context, normative notions of family structure, such as the "nuclear family", may not therefore reflect the reality of family life, experiences and functions, as described and articulated by families themselves; particularly those from minority or marginalized communities [1]. Despite this complexity and perpetual change, the importance of family for the experience of both interdependence and individual support and well-being remains constant. This is particularly the case for "families with complex needs", who experience both a "breadth" of "interrelated or interconnected" needs and a "depth" of "profound, severe, serious or intense needs" [2], and are therefore most reliant on services and support. This might include families affected by mental health needs, disability, caring responsibilities, migration, asylum seeking, crime, drug and alcohol misuse, and so on. The increasing complexity of family life, alongside the continued important and complex role played by family in supporting members with particular needs, poses a range of challenges for services seeking to engage with families, particularly those with complex needs. For family-focused services to deliver effectively, the complexity of family roles, functions, and compositions therefore need to be examined and understood. Failure to recognize the structure, role and function of various family relationships may lead to ineffective service provision or a resistance to engage in support by the family. However, in sharp contrast to this complexity and fluidity of experience, Murray and Barnes [3] argue that "family" is a taken-for-granted and narrowly defined concept within policy documentation in the UK, and highlight the importance of "exploring normative assumptions about family" that
INTERNATIONAL SYMPOSIUM ON CURRENT PROGRESS IN MATHEMATICS AND SCIENCES 2016 (ISCPMS 2016): Proceedings of the 2nd International Symposium on Current Progress in Mathematics and Sciences 2016, 2017
Extreme Value Theory (EVT) has emerged as one of the most important statistical theories for the ... more Extreme Value Theory (EVT) has emerged as one of the most important statistical theories for the applied sciences. EVT provides a firm theoretical foundation for building a statistical model describing extreme events. The feature that distinguish extreme value analysis than other statistical analysis is the ability to quantify the behavior of unusually large values even when those values are scarce. One of the key results from EVT is the ability to estimate the distribution of maximum value, that usually called as maxima, using the asymptotic argument. In order to build such models, the Fisher-Tippett theorem which specifies the form of the limit distribution for transformed maxima will be greatly used. Furthermore, it can be shown that there are only three families of possible limit laws for distribution of maxima, which are the Gumbel, Frechet, and Weibull distributions. These three distributions can be expressed in a single distribution function called the generalized extreme value (GEV) distribution.
Journal of physics, 2021
RCC which stands for regression correlation coefficient is an alternative measure strength of pre... more RCC which stands for regression correlation coefficient is an alternative measure strength of predictive that can be applied to the GLM model in which the distribution of response variable is not only normal. The RCC is constructed based on the definition of correlation coefficient by using generalized linear model (GLM). So, the RCC can be defined as a value that states the strength of the relationship between response variable Y and its conditional expectation given predictor variables E[Y| X ]. The RCC is one measure of predictable power that can satisfies the property like applicability, interpretability, consistency, and affinity. In general, the explicit form of RCC on GLM is difficult to find. However, when RCC is applied to the Poisson regression model and the predictor variables are assumed to be a normal multivariate distribution, an explicit form is found. This explicit form still contains the unknown parameters derived from the Poisson regression model. Therefore, we need to find an estimate of these parameters to obtain an estimator from the RCC. The Poisson regression model which still contains the unknown parameters are estimated using maximum likelihood method. Application of regression correlation coefficient is done in case of maternal mortality rate in Central Java Province in 2015.
Journal of physics, Mar 1, 2021
Residential property or home is one of the basic human needs as a shelter, to be able to continue... more Residential property or home is one of the basic human needs as a shelter, to be able to continue to live. As the population in Indonesia increases, express the need for residential property will also increase. The movement of residential property prices in Indonesia can be observed from the Residential Property Price Index (RPPI) issued by Bank Indonesia. Residential property prices can be influenced by several factors. For example in economic, the movement of RPPI can be influenced by several factors such as inflation, Composite Stock Price Index (RPPI) and loan interest rates. Forecasting the RPPI can be done with times series modeling. Modeling RPPI using three influencing variables requires a multivariate time series model. The multi-input transfer function model is a multivariate model that can be used in modeling RPPI. In the multi-input transfer function model there is an output series (y t ) which is the RPPI which is estimated to be influenced by several input series (x jt ), which are inflation, IHSG and loan interest rates. Based on the model obtained in this study, it can be seen that the prediction of RPPI at the t-time is influenced by the amount of inflation in the previous two months until previous five months, influence by loan interest rates without delay so that it is influenced by the t-time until previous three months, influenced by the IHSG at t-time until previous four months, and also influenced by itself at one month before to four months before.
Journal of physics, May 1, 2019
Forum Statistika dan Komputasi, Feb 26, 2021
Poisson distribution is a common distribution for modelling count data with assumption mean and v... more Poisson distribution is a common distribution for modelling count data with assumption mean and variance has the same value (equidispersion). In fact, most of the count data have mean that is smaller than variance (overdispersion) and Poisson distribution cannot be used for modelling this kind of data. Thus, several alternative distributions have been introduced to solve this problem. One of them is Shanker distribution that only has one parameter. Since Shanker distribution is continuous distribution, it cannot be used for modelling count data. Therefore, a new distribution is offered that is Poisson-Shanker distribution. Poisson-Shanker distribution is obtained by mixing Poisson and Shanker distribution, with Shanker distribution as the mixing distribution. The result is a mixture distribution that has one parameter and can be used for modelling overdispersion count data. In this paper, we obtain that Poisson-Shanker distribution has several properties are unimodal, overdispersion, increasing hazard rate, and right skew. The first four raw moments and central moments have been obtained. Maximum likelihood is a method that is used to estimate the parameter, and the solution can be done using numerical iterations. A real data set is used to illustrate the proposed distribution. The characteristics of the Poisson-Shanker distribution parameter is also obtained by numerical simulation with several variations in parameter values and sample size. The result is MSE and bias of the
Journal of physics, 2021
A risk is the possibility of undesirable events happening in the future. A good risk measure is n... more A risk is the possibility of undesirable events happening in the future. A good risk measure is needed to quantify the risks one faces. Some well-known risk measures include the Value-at-Risk (VaR) and the Expected Shortfall (ES). VaR measures the lower bound for big losses in a loss distribution tail, while ES measures the average of losses surpassing the VaR. Unfortunately, there are drawbacks in using the stated risk measures, mainly that they do not provide information regarding the variability of losses in the distribution tail. This paper will introduce and explore Gini Shortfall (GS), a more comprehensive risk measure than VaR and ES. GS provides information on the variability of data in distribution tails measured with Tail-Gini functional, a tail variability measure based on the variability measure Gini Mean Difference or Gini functional. Another superiority of GS is that under certain conditions, it can satisfy the four criteria of coherency. A coherent risk measure is useful for companies or investors to determine the right business and investing strategies. This paper will also provide explicit formulas of GS for some loss-related distributions, namely the exponential, Pareto, and logistic distributions. These formulas are then applied to calculate risks from actual stock data.
Journal of physics, 2020
Vehicle insurance companies in many countries use the Bonus-Malus System to determine the policyh... more Vehicle insurance companies in many countries use the Bonus-Malus System to determine the policyholder's net premium. The determination of net premiums on the Bonus-Malus System is based solely on the frequency of claims and ignores the severity of claims. This is unfair to policyholders who have small claims. To overcome this problem, the net premium determination method in Bonus-Malus System was developed taking into account both the frequency and severity of claims. Frequency and severity be assumed to be independent. In determining the net premium, a posterior distribution of parameters of the frequency and severity distribution is required. In the case of frequency and severity independent, the determination of the posterior distribution for frequency and severity is performed separately. This thesis discusses the determination of net premium based on frequency distribution and severity distribution for frequency and severity independent.
AIP Conference Proceedings
The most commonly used time series model is the discrete time series which assumes the variables ... more The most commonly used time series model is the discrete time series which assumes the variables being tested are continuous and produce continuous values. Whereas in many applications, a discrete time series model is needed to handle discrete variables and produce discrete values as well. The time series model that handles count or non-negative integer data is the Integer-valued Autoregressive model with the pth-order or INAR(p). This model is built with binomial thinning operator which implements probabilistic operations with discrete distribution that are suitable to model count data such as Poisson and Binomial. Model parameters will be estimated using the Yule-Walker method. In this research, we will discuss and describe the characteristics of the INAR(p) model using the binomial thinning operator. The INAR(p) specification follows the Autoregressive model with the pth order, AR(p). Forecasting in INAR(p) uses median forecasting by calculating the conditional probability of each possible non-negative integer value, then selecting a forecast value with a cumulative conditional probability greater than 0.5. The INAR(p) time series model will be applied to the 115 simulated data with non-negative integer values.
Journal of Physics: Conference Series, 2021
The failure-time distribution is used to describe the life of a device, material, or structure. O... more The failure-time distribution is used to describe the life of a device, material, or structure. One of the most commonly used distributions to analyze failure-time data is Weibull distribution. The Weibull distribution is very flexible for modeling varying types of failure rates data with constant, increasing and decreasing shapes, but cannot be used for modeling data with unimodal failure rates. Because of this, we developed the Weibull distribution using compounding method to produce a Weibull-Geometric distribution. The Weibull-Geometric distribution is useful for modeling failure rates data with a unimodal shape. In studies of failure, time to failure is frequently measured in the number of cycles to failure or shocks to failure and become a discrete random variable. Hence, this paper discusses the formation of distribution that more appropriate to modeling discrete failure data. The discrete distribution is obtained by discretizing the continuous Weibull-Geometric distribution....
Journal of Physics: Conference Series, 2021
Duration and convexity are two important factors in interest rate immunization. These two factors... more Duration and convexity are two important factors in interest rate immunization. These two factors used to be very closely related to financial assets immunization towards change in interest rate, but some of the latest studies had applied the concept of duration and immunization in terms of mortality immunization. This paper examines 24 mortality immunization strategies that applied the duration and convexity concept in insurance portfolios which consist of life insurance and annuity products. The outcome of this study is the optimal proportion of the life insurance and annuity products for life insurance companies in Indonesia. Numeric simulations had been done using Indonesia Mortality Table 2011 (TMI-2011) to obtain the value and characteristics of the optimal proportion for two portfolios which is affected by several factors, namely the implemented strategy, mortality model, mortality model change type, age of policy holder, year of application, payment paying period, and term p...
Journal of Physics: Conference Series, 2021
Risk is a probability of loss. In financial terms, loss can be interpreted as a possibility that ... more Risk is a probability of loss. In financial terms, loss can be interpreted as a possibility that an actual return on an investment will be lower than expected returns. Recent studies develop a new type of risk measures called credible value at risk (CrVaR). Credible value at risk is a model obtained by combining credibility theory and one of the most used risk measures, value at risk (VaR). Credibility theory is a model which gives a proper weight for both information and VaR is used to calculate maximum loss with the specific level of certainty and specific time frame. The combination of credibility theory and VaR is required to get a better value at risk estimation based on individual and group experiences. This paper discusses the model of credible value at risk, its parameter estimation, and focuses on the implementation of credible value at risk to predict the future rate of return from Indonesia’s stock market data.
PROCEEDINGS OF THE 6TH INTERNATIONAL SYMPOSIUM ON CURRENT PROGRESS IN MATHEMATICS AND SCIENCES 2020 (ISCPMS 2020), 2021
Mortality prediction is a crucial aspect for insurance and pension fund companies in deciding a s... more Mortality prediction is a crucial aspect for insurance and pension fund companies in deciding a suitable premium. The aim of this research is to discuss a cross-country (multi-population) mortality modeling in order to obtain a better mortality prediction. This modeling is based on the multi-dimensional Buhlmann credibility approach. The expansion in this research refers to mortality rate data taken from several countries. The Buhlmann credibility theory is generally used to predict the value of a random variable in a given period in the future. In this research, prediction for years to come was done using two strategies: Expanding Window and Moving Window. For every prediction in the upcoming period, both Expanding Window and Moving Window use prediction result values as additional data to build upon the prediction model for the next year; however, Moving Window also dismisses the oldest data. The model parameter is estimated with non-parametric approach. This model is then applied...
Journal of Physics: Conference Series, 2021
Based on the settlement period for insurance claim, insurance is divided into 2 types of business... more Based on the settlement period for insurance claim, insurance is divided into 2 types of business which are short-tail business (settlement period <1 year) and long-tail business (settlement period ≥1 year). In long-tail business, it is important for insurance company to have claim reserve in order to settle claims in the future. Claim reserve modelling is done using chain ladder method that is based on the trend of paid claims. Another method that is often used is Bornhuetter-Ferguson which is based on paid claims and also premium. In this paper, Benktander method that combines chain ladder and Bornhuetter-Ferguson using optimal credibility is introduced. Optimal credibility is obtained through minimum mean squared error and minimum variance. Benktander provides moderate reserve compared to chain ladder and Bornhuetter-Ferguson.
Advances in Intelligent Systems and Computing, 2014
In any classification problem, the dataset typically has a large number of features. However, not... more In any classification problem, the dataset typically has a large number of features. However, not all features are necessary to obtain a good classification performance because some of them are irrelevant and redundant. Therefore, classifiers with less number of features but with better classification accuracy are favored for ease of interpretation. In this work, particle swarm optimization algorithm along with logistic regression model is proposed. Additionally, the Bayesian information criterion (BIC) as a fitness function is proposed. The performance of different fitness functions is investigated and compared with BIC. The performance of the proposed method is evaluated based on a large number of different types of datasets. Experimental results using different types of datasets demonstrate the usefulness of our proposed method in significantly obtaining an improved classification performance with few features. Further, the results show that the proposed methods have a competitive performance comparing with other existing fitness functions.
Journal of physics, 2021
Burr distribution is Burr Type XII distribution which is one among the twelve types of the contin... more Burr distribution is Burr Type XII distribution which is one among the twelve types of the continuous distributions in Burr system. It has two positive shape parameters, namely k and c. It is implied from the probability density function which can be either decreasing or unimodal, and the hazard rate function which can be either decreasing or upside-down bathtub-shaped. The other distributional properties and the moment properties of Burr distribution will be discussed in more detail. By considering these properties, we will study its tail behaviour. To estimate the parameters k and c, the maximum likelihood method will be considered. Based on the properties of the data representing the remission time of bladder cancer patients, we infer that Burr distribution is suitable to model the data. The goodness-of-fit using the Kolmogorov–Smirnov test shows that Burr distribution fits well to the data.
Universitas Indonesia, 2012
Journal of physics, May 1, 2019
Family forms are many and varied, reflecting a myriad of understandings and influencing factors. ... more Family forms are many and varied, reflecting a myriad of understandings and influencing factors. In any given cultural context, normative notions of family structure, such as the "nuclear family", may not therefore reflect the reality of family life, experiences and functions, as described and articulated by families themselves; particularly those from minority or marginalized communities [1]. Despite this complexity and perpetual change, the importance of family for the experience of both interdependence and individual support and well-being remains constant. This is particularly the case for "families with complex needs", who experience both a "breadth" of "interrelated or interconnected" needs and a "depth" of "profound, severe, serious or intense needs" [2], and are therefore most reliant on services and support. This might include families affected by mental health needs, disability, caring responsibilities, migration, asylum seeking, crime, drug and alcohol misuse, and so on. The increasing complexity of family life, alongside the continued important and complex role played by family in supporting members with particular needs, poses a range of challenges for services seeking to engage with families, particularly those with complex needs. For family-focused services to deliver effectively, the complexity of family roles, functions, and compositions therefore need to be examined and understood. Failure to recognize the structure, role and function of various family relationships may lead to ineffective service provision or a resistance to engage in support by the family. However, in sharp contrast to this complexity and fluidity of experience, Murray and Barnes [3] argue that "family" is a taken-for-granted and narrowly defined concept within policy documentation in the UK, and highlight the importance of "exploring normative assumptions about family" that
INTERNATIONAL SYMPOSIUM ON CURRENT PROGRESS IN MATHEMATICS AND SCIENCES 2016 (ISCPMS 2016): Proceedings of the 2nd International Symposium on Current Progress in Mathematics and Sciences 2016, 2017
Extreme Value Theory (EVT) has emerged as one of the most important statistical theories for the ... more Extreme Value Theory (EVT) has emerged as one of the most important statistical theories for the applied sciences. EVT provides a firm theoretical foundation for building a statistical model describing extreme events. The feature that distinguish extreme value analysis than other statistical analysis is the ability to quantify the behavior of unusually large values even when those values are scarce. One of the key results from EVT is the ability to estimate the distribution of maximum value, that usually called as maxima, using the asymptotic argument. In order to build such models, the Fisher-Tippett theorem which specifies the form of the limit distribution for transformed maxima will be greatly used. Furthermore, it can be shown that there are only three families of possible limit laws for distribution of maxima, which are the Gumbel, Frechet, and Weibull distributions. These three distributions can be expressed in a single distribution function called the generalized extreme value (GEV) distribution.
Journal of physics, 2021
RCC which stands for regression correlation coefficient is an alternative measure strength of pre... more RCC which stands for regression correlation coefficient is an alternative measure strength of predictive that can be applied to the GLM model in which the distribution of response variable is not only normal. The RCC is constructed based on the definition of correlation coefficient by using generalized linear model (GLM). So, the RCC can be defined as a value that states the strength of the relationship between response variable Y and its conditional expectation given predictor variables E[Y| X ]. The RCC is one measure of predictable power that can satisfies the property like applicability, interpretability, consistency, and affinity. In general, the explicit form of RCC on GLM is difficult to find. However, when RCC is applied to the Poisson regression model and the predictor variables are assumed to be a normal multivariate distribution, an explicit form is found. This explicit form still contains the unknown parameters derived from the Poisson regression model. Therefore, we need to find an estimate of these parameters to obtain an estimator from the RCC. The Poisson regression model which still contains the unknown parameters are estimated using maximum likelihood method. Application of regression correlation coefficient is done in case of maternal mortality rate in Central Java Province in 2015.
Journal of physics, Mar 1, 2021
Residential property or home is one of the basic human needs as a shelter, to be able to continue... more Residential property or home is one of the basic human needs as a shelter, to be able to continue to live. As the population in Indonesia increases, express the need for residential property will also increase. The movement of residential property prices in Indonesia can be observed from the Residential Property Price Index (RPPI) issued by Bank Indonesia. Residential property prices can be influenced by several factors. For example in economic, the movement of RPPI can be influenced by several factors such as inflation, Composite Stock Price Index (RPPI) and loan interest rates. Forecasting the RPPI can be done with times series modeling. Modeling RPPI using three influencing variables requires a multivariate time series model. The multi-input transfer function model is a multivariate model that can be used in modeling RPPI. In the multi-input transfer function model there is an output series (y t ) which is the RPPI which is estimated to be influenced by several input series (x jt ), which are inflation, IHSG and loan interest rates. Based on the model obtained in this study, it can be seen that the prediction of RPPI at the t-time is influenced by the amount of inflation in the previous two months until previous five months, influence by loan interest rates without delay so that it is influenced by the t-time until previous three months, influenced by the IHSG at t-time until previous four months, and also influenced by itself at one month before to four months before.
Journal of physics, May 1, 2019
Forum Statistika dan Komputasi, Feb 26, 2021
Poisson distribution is a common distribution for modelling count data with assumption mean and v... more Poisson distribution is a common distribution for modelling count data with assumption mean and variance has the same value (equidispersion). In fact, most of the count data have mean that is smaller than variance (overdispersion) and Poisson distribution cannot be used for modelling this kind of data. Thus, several alternative distributions have been introduced to solve this problem. One of them is Shanker distribution that only has one parameter. Since Shanker distribution is continuous distribution, it cannot be used for modelling count data. Therefore, a new distribution is offered that is Poisson-Shanker distribution. Poisson-Shanker distribution is obtained by mixing Poisson and Shanker distribution, with Shanker distribution as the mixing distribution. The result is a mixture distribution that has one parameter and can be used for modelling overdispersion count data. In this paper, we obtain that Poisson-Shanker distribution has several properties are unimodal, overdispersion, increasing hazard rate, and right skew. The first four raw moments and central moments have been obtained. Maximum likelihood is a method that is used to estimate the parameter, and the solution can be done using numerical iterations. A real data set is used to illustrate the proposed distribution. The characteristics of the Poisson-Shanker distribution parameter is also obtained by numerical simulation with several variations in parameter values and sample size. The result is MSE and bias of the
Journal of physics, 2021
A risk is the possibility of undesirable events happening in the future. A good risk measure is n... more A risk is the possibility of undesirable events happening in the future. A good risk measure is needed to quantify the risks one faces. Some well-known risk measures include the Value-at-Risk (VaR) and the Expected Shortfall (ES). VaR measures the lower bound for big losses in a loss distribution tail, while ES measures the average of losses surpassing the VaR. Unfortunately, there are drawbacks in using the stated risk measures, mainly that they do not provide information regarding the variability of losses in the distribution tail. This paper will introduce and explore Gini Shortfall (GS), a more comprehensive risk measure than VaR and ES. GS provides information on the variability of data in distribution tails measured with Tail-Gini functional, a tail variability measure based on the variability measure Gini Mean Difference or Gini functional. Another superiority of GS is that under certain conditions, it can satisfy the four criteria of coherency. A coherent risk measure is useful for companies or investors to determine the right business and investing strategies. This paper will also provide explicit formulas of GS for some loss-related distributions, namely the exponential, Pareto, and logistic distributions. These formulas are then applied to calculate risks from actual stock data.
Journal of physics, 2020
Vehicle insurance companies in many countries use the Bonus-Malus System to determine the policyh... more Vehicle insurance companies in many countries use the Bonus-Malus System to determine the policyholder's net premium. The determination of net premiums on the Bonus-Malus System is based solely on the frequency of claims and ignores the severity of claims. This is unfair to policyholders who have small claims. To overcome this problem, the net premium determination method in Bonus-Malus System was developed taking into account both the frequency and severity of claims. Frequency and severity be assumed to be independent. In determining the net premium, a posterior distribution of parameters of the frequency and severity distribution is required. In the case of frequency and severity independent, the determination of the posterior distribution for frequency and severity is performed separately. This thesis discusses the determination of net premium based on frequency distribution and severity distribution for frequency and severity independent.
AIP Conference Proceedings
The most commonly used time series model is the discrete time series which assumes the variables ... more The most commonly used time series model is the discrete time series which assumes the variables being tested are continuous and produce continuous values. Whereas in many applications, a discrete time series model is needed to handle discrete variables and produce discrete values as well. The time series model that handles count or non-negative integer data is the Integer-valued Autoregressive model with the pth-order or INAR(p). This model is built with binomial thinning operator which implements probabilistic operations with discrete distribution that are suitable to model count data such as Poisson and Binomial. Model parameters will be estimated using the Yule-Walker method. In this research, we will discuss and describe the characteristics of the INAR(p) model using the binomial thinning operator. The INAR(p) specification follows the Autoregressive model with the pth order, AR(p). Forecasting in INAR(p) uses median forecasting by calculating the conditional probability of each possible non-negative integer value, then selecting a forecast value with a cumulative conditional probability greater than 0.5. The INAR(p) time series model will be applied to the 115 simulated data with non-negative integer values.
Journal of Physics: Conference Series, 2021
The failure-time distribution is used to describe the life of a device, material, or structure. O... more The failure-time distribution is used to describe the life of a device, material, or structure. One of the most commonly used distributions to analyze failure-time data is Weibull distribution. The Weibull distribution is very flexible for modeling varying types of failure rates data with constant, increasing and decreasing shapes, but cannot be used for modeling data with unimodal failure rates. Because of this, we developed the Weibull distribution using compounding method to produce a Weibull-Geometric distribution. The Weibull-Geometric distribution is useful for modeling failure rates data with a unimodal shape. In studies of failure, time to failure is frequently measured in the number of cycles to failure or shocks to failure and become a discrete random variable. Hence, this paper discusses the formation of distribution that more appropriate to modeling discrete failure data. The discrete distribution is obtained by discretizing the continuous Weibull-Geometric distribution....
Journal of Physics: Conference Series, 2021
Duration and convexity are two important factors in interest rate immunization. These two factors... more Duration and convexity are two important factors in interest rate immunization. These two factors used to be very closely related to financial assets immunization towards change in interest rate, but some of the latest studies had applied the concept of duration and immunization in terms of mortality immunization. This paper examines 24 mortality immunization strategies that applied the duration and convexity concept in insurance portfolios which consist of life insurance and annuity products. The outcome of this study is the optimal proportion of the life insurance and annuity products for life insurance companies in Indonesia. Numeric simulations had been done using Indonesia Mortality Table 2011 (TMI-2011) to obtain the value and characteristics of the optimal proportion for two portfolios which is affected by several factors, namely the implemented strategy, mortality model, mortality model change type, age of policy holder, year of application, payment paying period, and term p...
Journal of Physics: Conference Series, 2021
Risk is a probability of loss. In financial terms, loss can be interpreted as a possibility that ... more Risk is a probability of loss. In financial terms, loss can be interpreted as a possibility that an actual return on an investment will be lower than expected returns. Recent studies develop a new type of risk measures called credible value at risk (CrVaR). Credible value at risk is a model obtained by combining credibility theory and one of the most used risk measures, value at risk (VaR). Credibility theory is a model which gives a proper weight for both information and VaR is used to calculate maximum loss with the specific level of certainty and specific time frame. The combination of credibility theory and VaR is required to get a better value at risk estimation based on individual and group experiences. This paper discusses the model of credible value at risk, its parameter estimation, and focuses on the implementation of credible value at risk to predict the future rate of return from Indonesia’s stock market data.
PROCEEDINGS OF THE 6TH INTERNATIONAL SYMPOSIUM ON CURRENT PROGRESS IN MATHEMATICS AND SCIENCES 2020 (ISCPMS 2020), 2021
Mortality prediction is a crucial aspect for insurance and pension fund companies in deciding a s... more Mortality prediction is a crucial aspect for insurance and pension fund companies in deciding a suitable premium. The aim of this research is to discuss a cross-country (multi-population) mortality modeling in order to obtain a better mortality prediction. This modeling is based on the multi-dimensional Buhlmann credibility approach. The expansion in this research refers to mortality rate data taken from several countries. The Buhlmann credibility theory is generally used to predict the value of a random variable in a given period in the future. In this research, prediction for years to come was done using two strategies: Expanding Window and Moving Window. For every prediction in the upcoming period, both Expanding Window and Moving Window use prediction result values as additional data to build upon the prediction model for the next year; however, Moving Window also dismisses the oldest data. The model parameter is estimated with non-parametric approach. This model is then applied...
Journal of Physics: Conference Series, 2021
Based on the settlement period for insurance claim, insurance is divided into 2 types of business... more Based on the settlement period for insurance claim, insurance is divided into 2 types of business which are short-tail business (settlement period <1 year) and long-tail business (settlement period ≥1 year). In long-tail business, it is important for insurance company to have claim reserve in order to settle claims in the future. Claim reserve modelling is done using chain ladder method that is based on the trend of paid claims. Another method that is often used is Bornhuetter-Ferguson which is based on paid claims and also premium. In this paper, Benktander method that combines chain ladder and Bornhuetter-Ferguson using optimal credibility is introduced. Optimal credibility is obtained through minimum mean squared error and minimum variance. Benktander provides moderate reserve compared to chain ladder and Bornhuetter-Ferguson.
Advances in Intelligent Systems and Computing, 2014
In any classification problem, the dataset typically has a large number of features. However, not... more In any classification problem, the dataset typically has a large number of features. However, not all features are necessary to obtain a good classification performance because some of them are irrelevant and redundant. Therefore, classifiers with less number of features but with better classification accuracy are favored for ease of interpretation. In this work, particle swarm optimization algorithm along with logistic regression model is proposed. Additionally, the Bayesian information criterion (BIC) as a fitness function is proposed. The performance of different fitness functions is investigated and compared with BIC. The performance of the proposed method is evaluated based on a large number of different types of datasets. Experimental results using different types of datasets demonstrate the usefulness of our proposed method in significantly obtaining an improved classification performance with few features. Further, the results show that the proposed methods have a competitive performance comparing with other existing fitness functions.
Journal of physics, 2021
Burr distribution is Burr Type XII distribution which is one among the twelve types of the contin... more Burr distribution is Burr Type XII distribution which is one among the twelve types of the continuous distributions in Burr system. It has two positive shape parameters, namely k and c. It is implied from the probability density function which can be either decreasing or unimodal, and the hazard rate function which can be either decreasing or upside-down bathtub-shaped. The other distributional properties and the moment properties of Burr distribution will be discussed in more detail. By considering these properties, we will study its tail behaviour. To estimate the parameters k and c, the maximum likelihood method will be considered. Based on the properties of the data representing the remission time of bladder cancer patients, we infer that Burr distribution is suitable to model the data. The goodness-of-fit using the Kolmogorov–Smirnov test shows that Burr distribution fits well to the data.
Universitas Indonesia, 2012