Alternative balanced scorecards built from paradigm models in strategic HRM and employment/industrial relations and used to measure the state of employment relations and HR system performance across U.S. workplaces

This paper constructs alternative balanced scorecards based on high‐performance work system (HPWS) and employment relations system (ERS) models. The models are depicted and compared in diagrams and used as framework skeletons for building separate HPWS and ERS scorecards, intended to provide a detailed data picture of the operational health and performance of an organization's employment/HR system and its operations, processes, and inputs/outputs. The scorecards are filled in with nationally representative data from 2,000+ U.S. workplaces using more than 50 employment/HR indicators, as reported by separate panels of managers and employees. The indicators for each workplace are aggregated into an overall HR/employment system score, ranked from low‐to‐high, and graphed as frequency distributions. These distributions provide a unique snapshot picture of the mean and dispersion of the state of employment relations and HR system performance for companies across the United State. They also reveal that “models matter” since the HPWS and ERS scorecards provide distinctly different evaluation assessments.     

Comparison of intraday probabilistic forecasting of solar irradiance using only endogenous data

Accurate solar forecasts are necessary to improve the integration of solar renewables into the energy grid. In recent years, numerous methods have been developed for predicting the solar irradiance or the output of solar renewables. By definition, a forecast is uncertain. Thus, the models developed predict the mean and the associated uncertainty. Comparisons are therefore necessary and useful for assessing the skill and accuracy of these new methods in the field of solar energy.The aim of this paper is to present a comparison of various models that provide probabilistic forecasts of the solar irradiance within a very strict framework. Indeed, we consider focusing on intraday forecasts, with lead times ranging from 1 to 6 h. The models selected use only endogenous inputs for generating the forecasts. In other words, the only inputs of the models are the past solar irradiance data. In this context, the most common way of generating the forecasts is to combine point forecasting methods with probabilistic approaches in order to provide prediction intervals for the solar irradiance forecasts. For this task, we selected from the literature three point forecasting models (recursive autoregressive and moving average (ARMA), coupled autoregressive and dynamical system (CARDS), and neural network (NN)), and seven methods for assessing the distribution of their error (linear model in quantile regression (LMQR), weighted quantile regression (WQR), quantile regression neural network (QRNN), recursive generalized autoregressive conditional heteroskedasticity (GARCHrls), sieve bootstrap (SB), quantile regression forest (QRF), and gradient boosting decision trees (GBDT)), leading to a comparison of 20 combinations of models.None of the model combinations clearly outperform the others; nevertheless, some trends emerge from the comparison. First, the use of the clear sky index ensures the accuracy of the forecasts. This derived parameter permits time series to be deseasonalized with missing data, and is also a good explanatory variable of the distribution of the forecasting errors. Second, regardless of the point forecasting method used, linear models in quantile regression, weighted quantile regression and gradient boosting decision trees are able to forecast the prediction intervals accurately.     

江苏省大学生就业率预测分析

利用灰色系统理论建立了大学生就业率动态预测模型,结合江苏省就业率的统计数据对2003—2008年的就业率进行了模拟,并对结果进行了精度检验,结果表明,模型模拟精度高。最后,用此模型对江苏省未来5年的就业率进行了预测,并对就业形势进行了分析。     

Nonparametric Employment Subcenter Identification

A two-stage procedure is proposed for identifying urban employment subcenters. The first stage identifies candidate subcenters as significant positive residuals in a smoothed employment density function. Subcenters are those sites that provide significant explanatory power in the second-stage, semiparametric employment density function estimation. The procedure can be applied to either aggregated or disaggregated data, does not require detailed knowledge of the study area, and is easily reproducible by other researchers. Results are presented for five previously studied cities—Chicago, Dallas, Houston, Los Angeles, and San Francisco—and a new one, New Orleans.     

Comparison of methods for long-range assessment in education

There are many problems with attempting to assess the long-range impact of educational innovations or programs. The two fundamental components upon which most long-term educational evaluations are based are the use of standardized tests to measure cognitive growth and methodologies which project normal growth. Long-range educational evaluations are then based upon this notion of normal growth.

This study examines the accuracy of various student test score projection techniques-idiographic analysis, standard score method, regression analysis and a differential equation growth model. The study found that even though estimates for its parameters were derived from a small sample, the differential equation growth model was generally more accurate for long-range projection than the other models and approximately the same accuracy for short-term projections as regression. Idiographic analysis and the standard score method were generally poor in terms of accuracy of prediction, especially for extended time periods.

The study concludes with an overview of the problems of projecting student test scores and suggests a stochastic procedure based upon Bayesian analysis as a more realistic probabilistic projection technique for use in educational policy formulation—especially since this procedure is more compatible with expected value and benefit cost analysis which are now becoming an integral part of comprehensive instructional evaluation.     

On the selection of forecasting models

It is standard in applied work to select forecasting models by ranking candidate models by their prediction mean squared error (PMSE) in simulated out-of-sample (SOOS) forecasts. Alternatively, forecast models may be selected using information criteria (IC). We compare the asymptotic and finite-sample properties of these methods in terms of their ability to mimimize the true out-of-sample PMSE, allowing for possible misspecification of the forecast models under consideration. We show that under suitable conditions the IC method will be consistent for the best approximating model among the candidate models. In contrast, under standard assumptions the SOOS method, whether based on recursive or rolling regressions, will select overparameterized models with positive probability, resulting in excessive finite-sample PMSEs.     

Top-down strategies based on adaptive fuzzy rule-based systems for daily time series forecasting

This paper presents a data-driven approach applied to the long term prediction of daily time series in the Neural Forecasting Competition. The proposal comprises the use of adaptive fuzzy rule-based systems in a top-down modeling framework. Therefore, daily samples are aggregated to build weekly time series, and consequently, model optimization is performed in a top-down framework, thus reducing the forecast horizon from 56 to 8 steps ahead. Two different disaggregation procedures are evaluated: the historical and daily top-down approaches. Data pre-processing and input selection are carried out prior to the model adjustment. The prediction results are validated using multiple time series, as well as rolling origin evaluations with model re-calibration, and the results are compared with those obtained using daily models, allowing us to analyze the effectiveness of the top-down approach for longer forecast horizons.     

江苏省民办高职院校就业状况研究——基于灰色模型和回归模型的比较分析

本文尝试用灰色模型对相对数进行分解运算,并与传统的回归预测方法进行比较。一方面对高职学生的就业率进行预测,另一方面对灰色模型在相对数进行分解预算的准确性进行评价。     

Treating and Pruning: New approaches to forecasting model selection and combination using prediction intervals

We propose a new way of selecting among model forms in automated exponential smoothing routines, consequently enhancing their predictive power. The procedure, here addressed as treating, operates by selectively subsetting the ensemble of competing models based on information from their prediction intervals. By the same token, we set forth a pruning strategy to improve the accuracy of both point forecasts and prediction intervals in forecast combination methods. The proposed approaches are respectively applied to automated exponential smoothing routines and Bagging algorithms, to demonstrate their potential. An empirical experiment is conducted on a wide range of series from the M-Competitions. The results attest that the proposed approaches are simple, without requiring much additional computational cost, but capable of substantially improving forecasting accuracy for both point forecasts and prediction intervals, outperforming important benchmarks and recently developed forecast combination methods.     

The myth of purchasing professionals’ expertise. More evidence on whether computers can make better procurement decisions

In a previous experiment, we have shown that risk assessments of purchasing experts are certainly not better than that of subjects untrained in purchasing, and worse than the decisions made by formal models (J. Purchas. Supply Manage. 9 (2003) 191–198). Since both these results are rather counterintuitive, we conducted a series of experiments geared at replication and extension of these findings. These new experiments show that our previous results are robust, and reveal an additional finding that is both worrying and puzzling. It actually seems to be the case that for the purchasing decision tasks in our experiments, experts perform worse with growing experience. It therefore seems that, at least for the kinds of purchasing decisions under study, it does not make much sense to use expert judgments at all. However, we show that there is a way in which expert judgments can be used in combination with formal models to improve the predictive accuracy of purchasing predictions. In our case, superior predictions are made when we combine the prediction of a formal model with the prediction of the ‘average expert’, thereby combining the robust linear trends as encapsulated in the formal model with the more intuitive configural rules used by experts. We provide several explanations for this phenomenon.     

Estimating predictions,prediction errors and their standard deviations using constructed variables

Increasing use has been made of predictive tests for assessing model adequacy, but it is sometimes difficult to generate predictions and their standard errors in dynamic or simultaneous equation models. Following earlier suggestions by Salkever and Fuller, this paper shows how the requisite information may be obtained by the use of specially constructed variables in a regression framework. The main use of the method will be in those situations where prediction information is not available as a standard option in econometric packages.     

Forecasting China's foreign exchange reserves using dynamic model averaging: The roles of macroeconomic fundamentals,financial stress and economic uncertainty

We develop models for examining possible predictors of growth of China's foreign exchange reserves that embrace Chinese and global trade, financial and risk (uncertainty) factors. Specifically, by comparing with other alternative models, we show that the dynamic model averaging (DMA) and dynamic model selection (DMS) models outperform not only linear models (such as random walk, recursive OLS-AR(1) models, recursive OLS with all predictive variables models) but also the Bayesian model averaging (BMA) model for examining possible predictors of growth of those reserves. The DMS is the best overall across all forecast horizons. While some predictors matter more than others over the forecast horizons, there are few that stand the test of time. The US–China interest rate differential has a superior predictive power among the 13 predictors considered, followed by the nominal effective exchange rate and the interest rate spread for most of the forecast horizons. The relative predictive prowess of the oil and copper prices alternates, depending on the commodity cycles. Policy implications are also provided.     

Simple averaging of direct and recursive forecasts via partial pooling using machine learning

This article introduces the winning method at the M5 Accuracy competition. The presented method takes a simple manner of averaging the results of multiple base forecasting models that have been constructed via partial pooling of multi-level data. All base forecasting models of adopting direct or recursive multi-step forecasting methods are trained by the machine learning technique, LightGBM, from three different levels of data pools. At the competition, the simple averaging of the multiple direct and recursive forecasting models, called DRFAM, obtained the complementary effects between direct and recursive multi-step forecasting of the multi-level product sales to improve the accuracy and the robustness.     

The duality of bankruptcy process in Finland

Standard bankruptcy prediction methods lead to models weighted by the types of failure firms included in the estimation sample. These kinds of weighted models may lead to severe classification errors when they are applied to such types of failing (and non-failing) firms which are in the minority in the estimation sample (frequency effect). The purpose of this study is to present a bankruptcy prediction method based on identifying two different failure types, i.e. the solidity and liquidity bankruptcy firms, to avoid the frequency effect. Both of the types are depicted by a theoretical gambler's ruin model of its own to yield an approximation of failure probability separately for both types. These models are applied to the data of randomly selected Finnish bankrupt and non-bankrupt firms. A logistic regression model based on a set of financial variables is used as a benchmark model. Empirical results show that the resulting heavily solidity-weighted logistic model may lead to severe errors in classifying non-bankrupt firms. The present approach will avoid these kinds of error by separately evaluating the probability of the solidity and liquidity bankruptcy; the firm is not classified bankrupt as long as neither of the probabilities exceeds the critical value. This leads the present prediction method slightly to outperform the logistic model in the overall classification accuracy.     

The impact of obesity on employment

Using data from two rounds of the Health Survey for England I investigate the impact of obesity on employment. I use three approaches: a univariate probit model; propensity score matching; and IV regression using a recursive bivariate probit model. Conditional on a comprehensive set of covariates, the findings show that obesity has a statistically significant and negative effect on employment in both males and females. In males the endogeneity of obesity does not significantly affect the estimates, and the magnitude of effect is similar across the three methods. In females, failure to account for endogeneity leads to underestimation of the negative impact of obesity on employment.     

Long-range U.S. Manpower forecasts in retrospect: How accurate were we?

This paper compares and evaluates the accuracy of long-range occupational manpower forecasts made for 1980 in the early 1970s by the U.S. Bureau of Labor Statistics and by the author. The different assumptions and forecasting methodologies utilized are discussed, and the occupational forecasts are then compared to the actual 1980 employment data. The relative accuracy of the different sets of forecasts is assessed according to several different criteria, and the larger question of the usefulness of either set of forecasts is addressed. It is found that neither set of forecasts was clearly superior, that the accuracy of both sets of forecasts was generally poor, and that the projections for individual occupations were often so wide of the mark as to be of questionable usefulness for manpower planning and vocational guidance. The implications of these findings for manpower forecasting are discussed.The author is grateful to several referees for helpful comments on an earlier draft of this paper, but retains sole responsibility for the opinions expressed here and for any errors.     

A voted based random forests algorithm for smart grid distribution network faults prediction

ABSTRACT

In this paper, we focus on fault prediction in the smart distribution network. modified version of voted random forest algorithm (VRF) is proposed for enhancing the predicting accuracy of the faults. We change the decision process by redesigning the voting algorithm by introducing multiple SVM models for voting model training. Based on the trained models, a simple NSGA algorithm is applied to find the best voting model. Results showed that the new algorithm could improve the accuracy and recall rate of the fault prediction, especially for the recall rate of the negative samples.     

Structural information in recursive VAR orderings

This paper investigates structural models that will permit a Cholesky decomposition of the covariance matrix of VAR residuals to identify some structural impulse response functions. Cholesky decompositions are found to be useful identification tools for the set of partially recursive structural models. A partially recursive structure is defined as any block recursive system where the equations in one block can be recursively ordered and where the structural shocks are uncorrelated. Using this class of models, we derive necessary and sufficient conditions for the moving average representation from a Cholesky decomposition to identify structure. The paper concludes by discussing implications of these results for empirical research.     

Forecasting exchange rates using feedforward and recurrent neural networks

In this paper we investigate the out-of-sample forecasting ability of feedforward and recurrent neural networks based on empirical foreign exchange rate data. A two-step procedure is proposed to construct suitable networks, in which networks are selected based on the predictive stochastic complexity (PSC) criterion, and the selected networks are estimated using both recursive Newton algorithms and the method of nonlinear least squares. Our results show that PSC is a sensible criterion for selecting networks and for certain exchange rate series, some selected network models have significant market timing ability and/or significantly lower out-of-sample mean squared prediction error relative to the random walk model.     

The statistical analysis of overidentified linear recursive models

This paper examines identification and estimation in recursive linear models. After developing the main result on identification of recursive models, the paper considers estimation in models subject to overidentifying constraints. A particularly simple, but quite general and efficient, approach to estimating constrained recursive models is developed.