Dynamic mortality factor model with conditional heteroskedasticity

Forecasting the US term structure of interest rates using a macroeconomic smooth dynamic factor model

Dynamic Factor Models with Clustered Loadings: Forecasting Education Flows using Unemployment Data

Dynamic Factor Models with Control Variables

In the last decade, most studies on the interaction between the financial sector and economic growth has focused mainly on the banks and the stock market. Recently, growing attention has shifted to the interaction between the non-bank financial intermediaries (NBFIs) such as the insurance companies and economic growth. In this study, we examined the impact of insurance contributions on economic growth in Nigeria over a twenty seven year period, between 1981 and 2008 using a dynamic factor model. The proposed technique described a number of methods designed to analyze a functional relationship between the volume of insurance contribution and economic growth in terms of underlying but unobservable random quantities called factors. The factor loadings indicate which common trend is related to which set of time series. The result obtained through this approach shows that the functional relationship between the volume of insurance contribution and economic growth in Nigeria is a first order vector autoregressive model. Key words: Insurance contribution, dynamic factor model, economic growth.

Theoretical models of contagion and spillovers allow for asset-specific shocks that can be directly transmitted from one asset to another, as well as indirectly transmitted across uncorrelated assets through some intermediary mechanism. Standard multivariate Generalized Autoregressive Conditional Heteroskedasticity (GARCH) models, however, provide estimates of volatilities and correlations based only on the direct transmission of shocks across assets. As such, spillover effects via an intermediary asset or market are not considered. In this article, a multivariate GARCH model is constructed that provides estimates of volatilities and correlations based on both directly and indirectly transmitted shocks. The model is applied to exchange rate and equity returns data. The results suggest that if a spillover component is observed in the data, the spillover augmented models provide significantly different volatility estimates compared to standard multivariate GARCH models.

Improved index insurance design and yield estimation using a dynamic factor forecasting approach

The tiltrotor drive system is a complex and coupled system with different torsional dynamics compared to conventional rotorcraft. It is designed to guarantee safety and reliability and tested for the effects of its dynamics and loads on component sizing. The epicyclic gearing system, the final reduction stage in most aircraft gearboxes, is most impacted by such dynamics. The components of this subsystem must withstand all peak loads due to novel control strategies and torsional dynamics. In this paper, a gear contact analysis model is coupled with a validated torsional dynamics model of a tiltrotor interconnected drive system. The coupled model predicts component-level design loads in the interconnected drive system originating from its torsional dynamic behavior. An accurate gear macro and micro geometry is generated in the gear contact model which is used to derive the time varying mesh stiffnesses at the sub-component interfaces. The variable mesh stiffnesses are applied in the gear dynamic model to simulate the dynamic responses of the gear system under load. Finally, a dynamic overload factor relating the peak dynamic and static loads is analytically assessed, and it is shown to exceed conventional predictions in the case of an asymmetrically loaded interconnected drive system. This dynamic factor can be implemented to inform design decisions during the development of future multi-rotor VTOL aircrafts to mitigate component damage and, in turn, increase the safety and extend the lifespan of critical components.

This research aims to predict the mortality rates of Malaysian sub-populations in thirteen different states using the Three-Way Lee-Carter model: the extended version of the Lee-Carter model. This model considers the coherence characteristic between subpopulations as it allows combination estimates of the three variables: i) time, ii) age and iii) different subpopulations, which is useful to assess the magnitude of mortality differential between subpopulations. The Three-Way Lee-Carter model has never been tested using Malaysian mortality rates, particularly to forecast mortality of subpopulations in different states. Hence this research predicts the thirteen-state populations' mortality rates and compares the model's performance to the original Lee-Carter model. The collected data includes mortality rates by age, gender, and states obtained from the Department of Statistics Malaysia from 1991 until 2018. Results show that the Three-Way Lee-Carter model successfully fits into the Malaysia data better than the original Lee-Carter model. In terms of out-sample forecasts, the Three-Way Lee-Carter model outperforms the Lee-Carter model for eight subpopulations including for Kelantan males, Perlis males, Sabah males, Terengganu males, Perak females, Perlis females, Sarawak females and Selangor females.

The standard state space model treats observations as imprecise measurement of the Markovian states. Our flexible model handles the states and observations symmetrically, which are simultaneously determined by past observations and up to first‐lagged states. The only distinction between the states and observations is the observability. When it is applied to the autoregressive moving average, dynamic factor and stochastic volatility models, the state space form is both parsimonious and intuitive, for low‐dimension states are constructed simply by stacking all the relevant but unobserved components in the structural model.

Modeling and forecasting human mortality are significant research topics in several disciplines because mortality rates are fundamental in planning and policy decisions. Among various techniques, Lee Carter (LC) model is one of the most popular stochastic method in human mortality modeling. The original LC model was fuzzified to eliminate the assumptions related with homoscedasticity. The existing fuzzy model makes use of ordinary least squares (OLS) technique, which prevents the model to capture the existing fluctuations in data. In this study, a revised version of fuzzy LC model utilizing singular value decomposition (SVD) technique is proposed to overcome this issue. After modeling the mortality rates, their future values are forecasted by a modified first order fuzzy time series technique. For illustration purposes, proposed method is applied to mortality data of Finland. Numerical outputs show that proposed method is statistically better in modeling mortality compared to the existing fuzzy method. In addition, the modified fuzzy time series technique generates better forecasts than the original version.

Dynamics of processes in the chemical industries requires substantiation of new approaches to vocational training programs for future mechanical engineers. These scientific and practical innovations should affect the entire system of training, spreading, primarily on professionally oriented disciplines, leading among them for direction « Mechanical Engineering » specialty « chemical production equipment and construction materials » is « processes, machines and machine area » . The paper presents a system-factor approach to enhance the training of future mechanical engineers in the direction of « Engineering » specialty « chemical production equipment and building materials » , the discipline « processes, machines and machine industry » on the basis of system-factor organizing training tasks, which are purposefully reinforce this training for each conditionally selected for the leading function of work, professional category: « Technology » , « mechanics » , « designers » . So, the training of mechanical engineers in the direction of « Engineeing » specialty « chemical production equipment and building materials » should be aimed at the preparation of various experts said the work on the classification of professional groups. Each of the factors proposed system-factor approach to discipline « processes, machines and machine industry » to enhance the training of future mechanical engineers in the direction of « Engineering » is given and should be taken into account, but if necessary, subjected to the necessary adjustments. Conditional inclusion of subjective factors in the objective-subjective, and those – in an objective manner the static measurement of the factor model application system of learning tasks in the preparation of future mechanical engineers specialty « chemical production equipment and building materials » , the discipline « processes, machines and machine industry » . The dynamic factor model as the measurement system application training tasks forms a gradual taking into account the following factors each with a mandatory check the completeness of all previous records. Thus, the implementation of system-factor approach for future mechanical engineers in the direction of « Engineering » specialty « chemical production equipment and building materials » , the discipline « processes, machines and machine industry » to contribute to enhance their professional skills in the specialty for each conditionally selected for leading function of employment, Professional, « technology » , « mechanics » , « designers » . system-factor approach; a mechanical engineer; equipment; chemicals manufacturing and construction materials; engineering direction, objectives; factors, system-factor model

The Lee–Carter model could be considered as one of the most important mortality prediction models among stochastic models in the field of mortality. With the recent developments of machine learning and deep learning, many studies have applied deep learning approaches to time series mortality rate predictions, but most of them only focus on a comparison between the Long Short-Term Memory and the traditional models. In this study, three different recurrent neural networks, Long Short-Term Memory, Bidirectional Long Short-Term Memory, and Gated Recurrent Unit, are proposed for the task of mortality rate prediction. Different from the standard country level mortality rate comparison, this study compares the three deep learning models and the classic Lee–Carter model on nine divisions’ yearly mortality data by gender from 1966 to 2015 in the United States. With the out-of-sample testing, we found that the Gated Recurrent Unit model showed better average MAE and RMSE values than the Lee–Carter model on 72.2% (13/18) and 67.7% (12/18) of the database, respectively, while the same measure for the Long Short-Term Memory model and Bidirectional Long Short-Term Memory model are 50%/38.9% (MAE/RMSE) and 61.1%/61.1% (MAE/RMSE), respectively. If we consider forecasting accuracy, computing expense, and interpretability, the Lee–Carter model with ARIMA exhibits the best overall performance, but the recurrent neural networks could also be good candidates for mortality forecasting for divisions in the United States.

Cold-start problem is a key challenge in recommender system. In order to solve this problem, we proposed a dynamic latent factor model that combined with attributes of user and item. With the increase of interaction between user and recommender system, our model can automatically adjust its coefficient to achieve a more reasonable and more satisfactory recommendation. The experimental results show that the proposed method can not only alleviate the cold-start problem of new users and new items, but also improve the recommendation accuracy to some extent.

In the context of the exact dynamic common factor model, canonical correlations in a multivariate time series are used to identify the number of latent common factors. In this paper, we establish a relationship between canonical correlations and the autocovariance function of the factor process, in order to modify a pre-established statistical test to detect the number of common factors. In particular, the test power is increased. Additionally, we propose a procedure to identify a vector ARMA model for the factor process, which is based on the so-called simple and partial canonical autocorrelation functions. We illustrate the proposed methodology by means of some simulated examples and a real data application.

In this contribution we present a structure theory for generalized linear dynamic factor models. Generalized dynamic factor models have been proposed approximately a decade ago for modeling of high dimensional time series where the cross sectional dimension is of the same order of magnitude as the sample size. In these models the classical assumption for factor models, that the noise components are mutually uncorrelated, is relaxed by allowing for weak dependence. Structure theory turns out to be important for estimation and model selection. The results obtained heavily draw from linear system theory.The contribution consists of two main parts. In the first part we deal with “denoising”, i.e. with getting rid of the noise in the observations. In the second part we deal with constructing linear dynamic systems for the latent variables. Here an important result is the generic zerolessness of the transfer function relating the latent variables and the dynamic factors. This allows for modeling the latent variables by (singular) autoregressions which simplifies estimation.