Search results

The purpose of this study is to propose an effective unsupervised one-class-classifier (OCC) support vector machine (SVM)-based single multivariate control chart (OCC-SVM) to simultaneously monitor “location” and “scale” shifts of a manufacturing process.

The step-by-step approach to developing, implementing and fine-tuning the intrinsic parameters of the OCC-SVM chart is demonstrated based on simulation and two real-life case examples.

A comparative study, considering varied known and unknown response distributions, indicates that the OCC-SVM is highly effective in detecting process shifts of samples with individual observations. OCC-SVM chart also shows promising results for samples with a rational subgroup of observations. In addition, the results also indicate that the performance of OCC-SVM is unaffected by the small reference sample size.

The sample responses are considered identically distributed with no significant multivariate autocorrelation between sample observations.

The proposed easy-to-implement chart shows satisfactory performance to detect an out-of-control signal with known or unknown response distributions.

Various multivariate (e.g. parametric or nonparametric) control chart(s) are recommended to monitor the mean (e.g. location) and variance (e.g. scale) of multiple correlated responses in a manufacturing process. However, real-life implementation of a parametric control chart may be complex due to its restrictive response distribution assumptions. There is no evidence of work in the open literature that demonstrates the suitability of an unsupervised OCC-SVM chart to simultaneously monitor “location” and “scale” shifts of multivariate responses. Thus, a new efficient OCC-SVM single chart approach is proposed to address this gap to monitor a multivariate manufacturing process with unknown response distributions.

The purpose of the current study is to identify variables that, when integrated into the random effects parametric survival model, could be used to forecast the failure rate of small banks in the USA. A bank’s income production, efficiency and costs were taken into consideration when choosing the internal components. The breakout of the financial crisis, bank regulations that affect how the banking sector operates and the federal funds rate are the primary external variables.

This study uses the random effects parametric survival model to investigate the causes of small bank failures in the USA from 1996 to 2019. The study identifies several characteristics that failed banks frequently display. The main indications that may help to identify the elevated risk of small bank failures include the ROA, the cost of funds, the ratio of noninterest income to assets, the ratio of loan and lease losses to assets, noninterest expenses and core capital (leverage) ratio to assets. Economic disruptions, financial market distress and industry-based regulatory redress by the government exacerbate the financial distress borne by small banks.

The study revealed that a failed bank typically demonstrates a certain number of characteristics. The key factors that might assist identify which bank would be most likely to collapse include the cost of funding earning assets, the yield on earning assets, core Capital (leverage) ratio to assets, loan and lease loss provision to assets, noninterest expense and noninterest income to assets. Additionally, when a financial crisis occurs or the government changes regulations that could raise the cost of compliance for small banks, the likelihood that a bank will fail increases.

Models based on survival theories are more suitable when the authors examine bank failure as a unique event that happens gradually. The authors use a random effects parametric survival model to investigate the internal and external factors that may influence prospective small bank failure. This model has been developed and used in the medicinal research field. The authors do not choose the Cox proportional hazards model because it does not work well with panel data.

Equilibrium job search models allow for labor markets with homogeneous workers and firms to yield nondegenerate wage densities. However, the resulting wage densities do not accord well with empirical regularities. Accordingly, many extensions to the basic equilibrium search model have been considered (e.g., heterogeneity in productivity, heterogeneity in the value of leisure, etc.). It is increasingly common to use nonparametric forms for these extensions and, hence, researchers can obtain a perfect fit (in a kernel smoothed sense) between theoretical and empirical wage densities. This makes it difficult to carry out model comparison of different model extensions. In this paper, we first develop Bayesian parametric and nonparametric methods which are comparable to the existing non-Bayesian literature. We then show how Bayesian methods can be used to compare various nonparametric equilibrium search models in a statistically rigorous sense.

This paper gives a selective review on some recent developments of nonparametric methods in both continuous and discrete time finance, particularly in the areas of nonparametric estimation and testing of diffusion processes, nonparametric testing of parametric diffusion models, nonparametric pricing of derivatives, nonparametric estimation and hypothesis testing for nonlinear pricing kernel, and nonparametric predictability of asset returns. For each financial context, the paper discusses the suitable statistical concepts, models, and modeling procedures, as well as some of their applications to financial data. Their relative strengths and weaknesses are discussed. Much theoretical and empirical research is needed in this area, and more importantly, the paper points to several aspects that deserve further investigation.

The purpose of this paper is to present a nonparametric comparative study on the HCMS consumption of urban and rural households in China.

This paper applies the panel data for China's thirty provinces in 1991‐2007 as a sample and presents a local linear estimation to the nonparametric Working‐Leser panel data models on the HCMS consumption of the rural and urban households in China. The nonparametric Hausman test is applied to test the random effects specification against the fixed effects.

Both the parametric and nonparametric estimation of the Working‐Leser panel data models give us a similar result: that the HCMS commodity is quite elastic in both the rural and urban China. Nonparametric estimate also shows that the urban‐rural difference of the HCMS expenditure share in the total expenditure is mainly due to the large urban‐rural difference of total expenditure. Under the same total expenditure, the HCMS is more elastic in the urban than in the rural regions.

To decrease the urban‐rural difference in the HCMS consumption, the government should enhance the income or total expenditure level of the rural households, especially in the impoverished and remote areas in China. Urbanization plays a critical role in access to health care and can help make substantial changes in rural health care in China.

Compared to the parametric estimation, the nonparametric estimation gives us the added information that the expenditure elasticity of the HCMS consumption in China gradually declines as one moves up the per capita total expenditure distribution. The paper could make a contribution to the relatively thin literature on the Chinese medical and health consumption market.

This paper revisits the derivation and properties of the Allen-Uzawa and Morishima elasticities. Using a Swiss dataset, this paper empirically estimates various elasticities both in a dual and primal framework using a production theory open economy model and tests for linear homogenous technology. In addition to reporting elasticity at the mean, the standard practice in the literature, this paper also calculates nonparametric distribution of various elasticities. The paper aims to discuss these issues.

To assess the effect of price change on input, the paper estimates a translog cost function and to assess the effect of quantity change on price, the paper estimates the translog distance function using the data on Swiss economy. The paper estimates Allen-Uzawa and Morishima elasticity both under homogenous and non-homogenous technology using the Swiss dataset of one aggregate gross output and four inputs (resident labor, non-resident labor, imports, and capital) over 1950-1986. Elasticities are reported and compared at the mean as well as explored by looking at the range and nonparametric distribution.

This paper shows that constant returns to scale are easily rejected in this dataset and that the elasticities, both qualitatively and quantitatively, are very different under homogenous and non-homogenous technology. These elasticities can switch from complements to substitutes or vice versa when one moves away from the mean of the sample. The equality of the nonparametric elasticity distributions under homogenous vs non-homogenous technology is rejected in all cases except one.

This paper gives a clear derivation and interpretation of different elasticities as well as demonstrates using a dataset how to systematically go about empirically estimating these elasticities in a dual and primal framework. It shows that linear homogenous technology can be easily rejected and the elasticities, both quantitatively and qualitatively, are very different under homogenous and non-homogenous technology. This paper is also very valuable because it shows that the standard practice of reporting elasticity at the mean might not be adequate and there is a possibility that these elasticities can switch from complements to substitutes or vice versa when one moves away from the mean of the sample.

While the extant literature is replete with theoretical and empirical studies of value at risk (VaR) methods, only a few papers have applied the concept of VaR to quantify market risk in the context of agricultural finance. Furthermore, papers that have done so have largely relied on parametric methods to recover estimates of the VaR. The purpose of this paper is to assess extreme market risk on investment in three actively traded agricultural commodity futures.

A nonparametric Kernel method was implemented which accommodates fat tails and asymmetry of the portfolio return density as well as serial correlation of the data, to estimate market risk for investments in three actively traded agricultural futures contracts: corn, soybeans, and wheat. As a futures contract is a zero‐sum game, the VaR for both short and long sides of the market was computed.

It was found that wheat futures are riskier than either corn or soybeans futures over both periods considered in the study (2000‐2008 and 2006‐2008) and that all three commodities have experienced a sharp increase in market risk over the 2006‐2008 period, with VaR estimates 10‐43 percent higher than the long‐run estimates.

Research is based on cross‐sectional data and does not allow for dynamic assessment of expenditure elasticities.

This paper differs methodologically from previous applications of VaR in agricultural finance in that a nonparametric Kernel estimator was implemented which is exempt of misspecification risk, in the context of risk management of investment in agricultural futures contracts. The application is particularly relevant to grain elevator businesses which purchase grain from farmers on a forward contract basis and then turn to the futures markets to insure against falling prices.

The R environment for statistical computing and graphics (R Development Core Team, 2008) offers practitioners a rich set of statistical methods ranging from random number generation and optimization methods through regression, panel data, and time series methods, by way of illustration. The standard R distribution (base R) comes preloaded with a rich variety of functionality useful for applied econometricians. This functionality is enhanced by user-supplied packages made available via R servers that are mirrored around the world. Of interest in this chapter are methods for estimating nonparametric and semiparametric models. We summarize many of the facilities in R and consider some tools that might be of interest to those wishing to work with nonparametric methods who want to avoid resorting to programming in C or Fortran but need the speed of compiled code as opposed to interpreted code such as Gauss or Matlab by way of example. We encourage those working in the field to strongly consider implementing their methods in the R environment thereby making their work accessible to the widest possible audience via an open collaborative forum.

Economic conditions such as convexity, homogeneity, homotheticity, and monotonicity are all important assumptions or consequences of assumptions of economic functionals to be estimated. Recent research has seen a renewed interest in imposing constraints in nonparametric regression. We survey the available methods in the literature, discuss the challenges that present themselves when empirically implementing these methods, and extend an existing method to handle general nonlinear constraints. A heuristic discussion on the empirical implementation for methods that use sequential quadratic programming is provided for the reader, and simulated and empirical evidence on the distinction between constrained and unconstrained nonparametric regression surfaces is covered.