scholarly journals Calibrating the CreditRisk+ Model at Different Time Scales and in Presence of Temporal Autocorrelation †

Mathematics ◽  
2021 ◽  
Vol 9 (14) ◽  
pp. 1679
Author(s):  
Jacopo Giacomelli ◽  
Luca Passalacqua
Keyword(s):  
Time Series ◽  
Default Risk ◽  
Real Life ◽  
Statistical Error ◽  
Real Data ◽  
Sampling Period ◽  
Calibration Procedure ◽  
Industry Standards ◽  
Default Rate

The CreditRisk+ model is one of the industry standards for the valuation of default risk in credit loans portfolios. The calibration of CreditRisk+ requires, inter alia, the specification of the parameters describing the structure of dependence among default events. This work addresses the calibration of these parameters. In particular, we study the dependence of the calibration procedure on the sampling period of the default rate time series, that might be different from the time horizon onto which the model is used for forecasting, as it is often the case in real life applications. The case of autocorrelated time series and the role of the statistical error as a function of the time series period are also discussed. The findings of the proposed calibration technique are illustrated with the support of an application to real data.

Bayesian Structural Time Series

2020 ◽  
Vol 12 (1) ◽  
pp. 54-61
Author(s):  
Abdullah M. Almarashi ◽  
Khushnoor Khan
Keyword(s):  
Time Series ◽  
Stock Prices ◽  
Predictive Accuracy ◽  
Local Level ◽  
Real Life ◽  
Secondary Data ◽  
Real Data ◽  
Data Set ◽  
Structural Time Series ◽  
Structural Time

The current study focused on modeling times series using Bayesian Structural Time Series technique (BSTS) on a univariate data-set. Real-life secondary data from stock prices for flying cement covering a period of one year was used for analysis. Statistical results were based on simulation procedures using Kalman filter and Monte Carlo Markov Chain (MCMC). Though the current study involved stock prices data, the same approach can be applied to complex engineering process involving lead times. Results from the current study were compared with classical Autoregressive Integrated Moving Average (ARIMA) technique. For working out the Bayesian posterior sampling distributions BSTS package run with R software was used. Four BSTS models were used on a real data set to demonstrate the working of BSTS technique. The predictive accuracy for competing models was assessed using Forecasts plots and Mean Absolute Percent Error (MAPE). An easyto-follow approach was adopted so that both academicians and practitioners can easily replicate the mechanism. Findings from the study revealed that, for short-term forecasting, both ARIMA and BSTS are equally good but for long term forecasting, BSTS with local level is the most plausible option.

scholarly journals A teaching tool about the fickle p value and other statistical principles based on real-life data

2021 ◽  
Author(s):  
Salem Alawbathani ◽  
Mehreen Batool ◽  
Jan Fleckhaus ◽  
Sarkawt Hamad ◽  
Floyd Hassenrück ◽  
...  
Keyword(s):  
Statistical Analysis ◽  
Real Life ◽  
Real Data ◽  
P Value ◽  
Life Data ◽  
Real Life Data ◽  
Random Samples ◽  
Poor Understanding ◽  
The Impact

AbstractA poor understanding of statistical analysis has been proposed as a key reason for lack of replicability of many studies in experimental biomedicine. While several authors have demonstrated the fickleness of calculated p values based on simulations, we have experienced that such simulations are difficult to understand for many biomedical scientists and often do not lead to a sound understanding of the role of variability between random samples in statistical analysis. Therefore, we as trainees and trainers in a course of statistics for biomedical scientists have used real data from a large published study to develop a tool that allows scientists to directly experience the fickleness of p values. A tool based on a commonly used software package was developed that allows using random samples from real data. The tool is described and together with the underlying database is made available. The tool has been tested successfully in multiple other groups of biomedical scientists. It can also let trainees experience the impact of randomness, sample sizes and choice of specific statistical test on measured p values. We propose that live exercises based on real data will be more impactful in the training of biomedical scientists on statistical concepts.

scholarly journals Skew Generalized Normal Innovations for the AR(p) Process Endorsing Asymmetry

Symmetry ◽  
2020 ◽  
Vol 12 (8) ◽  
pp. 1253
Author(s):  
Ané Neethling ◽  
Johan Ferreira ◽  
Andriëtte Bekker ◽  
Mehrdad Naderi
Keyword(s):  
Time Series ◽  
Heavy Tails ◽  
Real Life ◽  
Value Added ◽  
Real Data ◽  
Behavior Simulation ◽  
Asymmetric Behavior ◽  
Normal Behavior ◽  
Ar Modeling ◽  
Conditional Maximum

The assumption of symmetry is often incorrect in real-life statistical modeling due to asymmetric behavior in the data. This implies a departure from the well-known assumption of normality defined for innovations in time series processes. In this paper, the autoregressive (AR) process of order p (i.e., the AR(p) process) is of particular interest using the skew generalized normal (SGN) distribution for the innovations, referred to hereafter as the ARSGN(p) process, to accommodate asymmetric behavior. This behavior presents itself by investigating some properties of the SGN distribution, which is a fundamental element for AR modeling of real data that exhibits non-normal behavior. Simulation studies illustrate the asymmetry and statistical properties of the conditional maximum likelihood (ML) parameters for the ARSGN(p) model. It is concluded that the ARSGN(p) model accounts well for time series processes exhibiting asymmetry, kurtosis, and heavy tails. Real time series datasets are analyzed, and the results of the ARSGN(p) model are compared to previously proposed models. The findings here state the effectiveness and viability of relaxing the normal assumption and the value added for considering the candidacy of the SGN for AR time series processes.

The role of parental self-regulation and household chaos in parent-toddler interactions: A time-series study.

2020 ◽  
Author(s):  
Sanne B. Geeraerts ◽  
Joyce Endendijk ◽  
Kirby Deater-Deckard ◽  
Jorg Huijding ◽  
Marike H. F. Deutz ◽  
...  
Keyword(s):  
Time Series ◽  
Self Regulation ◽  
Household Chaos ◽  
Time Series Study ◽  
Series Study

scholarly journals AnANN Model Trained on Regional Data in the Prediction of Particular Weather Conditions

2021 ◽  
Vol 11 (11) ◽  
pp. 4757
Author(s):  
Aleksandra Bączkiewicz ◽  
Jarosław Wątróbski ◽  
Wojciech Sałabun ◽  
Joanna Kołodziejczyk
Keyword(s):  
Atmospheric Pressure ◽  
Real Life ◽  
Real Data ◽  
Weather Conditions ◽  
Maximum Temperature ◽  
Air Masses ◽  
Weather Forecasts ◽  
Life Problems ◽  
The World ◽  
The City

Artificial Neural Networks (ANNs) have proven to be a powerful tool for solving a wide variety of real-life problems. The possibility of using them for forecasting phenomena occurring in nature, especially weather indicators, has been widely discussed. However, the various areas of the world differ in terms of their difficulty and ability in preparing accurate weather forecasts. Poland lies in a zone with a moderate transition climate, which is characterized by seasonality and the inflow of many types of air masses from different directions, which, combined with the compound terrain, causes climate variability and makes it difficult to accurately predict the weather. For this reason, it is necessary to adapt the model to the prediction of weather conditions and verify its effectiveness on real data. The principal aim of this study is to present the use of a regressive model based on a unidirectional multilayer neural network, also called a Multilayer Perceptron (MLP), to predict selected weather indicators for the city of Szczecin in Poland. The forecast of the model we implemented was effective in determining the daily parameters at 96% compliance with the actual measurements for the prediction of the minimum and maximum temperature for the next day and 83.27% for the prediction of atmospheric pressure.

A deep metric learning approach for weakly supervised loan default prediction1

2021 ◽  
pp. 1-13
Author(s):  
Kai Zhuang ◽  
Sen Wu ◽  
Xiaonan Gao
Keyword(s):  
Default Risk ◽  
Metric Learning ◽  
Real Life ◽  
Risk Scores ◽  
Loan Default ◽  
Default Prediction ◽  
Heterogeneous Features ◽  
Deep Metric Learning ◽  
Weakly Supervised ◽  
Decision Boundaries

To deal with the systematic risk of financial institutions and the rapid increasing of loan applications, it is becoming extremely important to automatically predict the default probability of a loan. However, this task is non-trivial due to the insufficient default samples, hard decision boundaries and numerous heterogeneous features. To the best of our knowledge, existing related researches fail in handling these three difficulties simultaneously. In this paper, we propose a weakly supervised loan default prediction model WEAKLOAN that systematically solves all these challenges based on deep metric learning. WEAKLOAN is composed of three key modules which are used for encoding loan features, learning evaluation metrics and calculating default risk scores. By doing so, WEAKLOAN can not only extract the features of a loan itself, but also model the hidden relationships in loan pairs. Extensive experiments on real-life datasets show that WEAKLOAN significantly outperforms all compared baselines even though the default loans for training are limited.

scholarly journals Intra-Colonial Viral Infections in Western Honey Bees (Apis mellifera)

2021 ◽  
Vol 9 (5) ◽  
pp. 1087
Author(s):  
Loreley Castelli ◽  
María Laura Genchi García ◽  
Anne Dalmon ◽  
Daniela Arredondo ◽  
Karina Antúnez ◽  
...  
Keyword(s):  
Apis Mellifera ◽  
Honey Bee ◽  
Viral Infections ◽  
Sampling Period ◽  
Virus Genome ◽  
The Individual ◽  
Bee Virus ◽  
Insight Into ◽  
Bee Viruses

RNA viruses play a significant role in the current high losses of pollinators. Although many studies have focused on the epidemiology of western honey bee (Apis mellifera) viruses at the colony level, the dynamics of virus infection within colonies remains poorly explored. In this study, the two main variants of the ubiquitous honey bee virus DWV as well as three major honey bee viruses (SBV, ABPV and BQCV) were analyzed from Varroa-destructor-parasitized pupae. More precisely, RT-qPCR was used to quantify and compare virus genome copies across honey bee pupae at the individual and subfamily levels (i.e., patrilines, sharing the same mother queen but with different drones as fathers). Additionally, virus genome copies were compared in cells parasitized by reproducing and non-reproducing mite foundresses to assess the role of this vector. Only DWV was detected in the samples, and the two variants of this virus significantly differed when comparing the sampling period, colonies and patrilines. Moreover, DWV-A and DWV-B exhibited different infection patterns, reflecting contrasting dynamics. Altogether, these results provide new insight into honey bee diseases and stress the need for more studies about the mechanisms of intra-colonial disease variation in social insects.

scholarly journals The Role of Assumptions in Ohlson Model Performance: Lessons for Improving Equity-Value Modeling

Mathematics ◽  
2021 ◽  
Vol 9 (5) ◽  
pp. 513
Author(s):  
Olga Fullana ◽  
Mariano González ◽  
David Toscano
Keyword(s):  
Time Series ◽  
Model Performance ◽  
Listed Firms ◽  
Valuation Model ◽  
Residual Income Valuation ◽  
Equity Value ◽  
Short Run ◽  
Time Series Approach ◽  
Linear Information

In this paper, we test whether the short-run econometric conditions for the basic assumptions of the Ohlson valuation model hold, and then we relate these results with the fulfillment of the short-run econometric conditions for this model to be effective. Better future modeling motivated us to analyze to what extent the assumptions involved in this seminal model are not good enough approximations to solve the firm valuation problem, causing poor model performance. The model is based on the well-known dividend discount model and the residual income valuation model, and it adds a linear information model, which is a time series model by nature. Therefore, we adopt the time series approach. In the presence of non-stationary variables, we focus our research on US-listed firms for which more than forty years of data with the required cointegration properties to use error correction models are available. The results show that the clean surplus relation assumption has no impact on model performance, while the unbiased accounting property assumption has an important effect on it. The results also emphasize the uselessness of forcing valuation models to match the value displacement property of dividends.

scholarly journals A Stochastic Kinetic Type Reactions Model for COVID-19

Mathematics ◽  
2021 ◽  
Vol 9 (11) ◽  
pp. 1221
Author(s):  
Giorgio Sonnino ◽  
Fernando Mora ◽  
Pasquale Nardone
Keyword(s):  
Real Data ◽  
Stochastic Noise ◽  
Mathematical Basis ◽  
Sis Model ◽  
Theoretical Predictions ◽  
Coronavirus Infection ◽  
Massive Distribution ◽  
Second Wave ◽  
Stochastic Sis Model

We propose two stochastic models for the Coronavirus pandemic. The statistical properties of the models, in particular the correlation functions and the probability density functions, were duly computed. Our models take into account the adoption of lockdown measures as well as the crucial role of hospitals and health care institutes. To accomplish this work we adopt a kinetic-type reaction approach where the modelling of the lockdown measures is obtained by introducing a new mathematical basis and the intensity of the stochastic noise is derived by statistical mechanics. We analysed two scenarios: the stochastic SIS-model (Susceptible ⇒ Infectious ⇒ Susceptible) and the stochastic SIS-model integrated with the action of the hospitals; both models take into account the lockdown measures. We show that, for the case of the stochastic SIS-model, once the lockdown measures are removed, the Coronavirus infection will start growing again. However, the combined contributions of lockdown measures with the action of hospitals and health institutes is able to contain and even to dampen the spread of the SARS-CoV-2 epidemic. This result may be used during a period of time when the massive distribution of vaccines in a given population is not yet feasible. We analysed data for USA and France. In the case of USA, we analysed the following situations: USA is subjected to the first wave of infection by Coronavirus and USA is in the second wave of SARS-CoV-2 infection. The agreement between theoretical predictions and real data confirms the validity of our approach.

Sign in / Sign up

Export Citation Format

Share Document