PARAMETRIC AND NON-PARAMETRIC COMBINATION MODEL TO ENHANCE OVERALL PERFORMANCE ON DEFAULT PREDICTION

The probability of default(PD) is the key element in the New Basel Capital Accord and the most essential factor to financial institutions' risk management.To obtain good PD estimation,practitioners and academics have put forward numerous default prediction models.However,how to use multiple models to enhance overall performance on default prediction remains untouched.In this paper,a parametric and non-parametric combination model is proposed.Firstly,binary logistic regression model(BLRM),support vector machine(SVM),and decision tree(DT) are used respectively to establish models with relatively stable and high performance.Secondly,in order to make further improvement to the overall performance,a combination model using the method of multiple discriminant analysis(MDA) is constructed.In this way,the coverage rate of the combination model is greatly improved,and the risk of miscarriage is effectively reduced.Lastly,the results of the combination model are analyzed by using the K-means clustering,and the clustering distribution is consistent with a normal distribution.The results show that the combination model based on parametric and non-parametric can effectively enhance the overall performance on default prediction.

Factors associated with crash severities in built-up areas along rural highways of Nevada:A case study of 11 towns

In 2014, 32,675 deaths were recorded in vehicle crashes within the United States. Out of these, 51% of the fatalities occurred in rural highways compared to 49% in urban highways. No specific crash data are available for the built-up areas along rural highways. Due to high fatalities in rural highways, it is important to identify the factors that cause the vehicle crashes. The main objective of this study is to determine the factors associated with se- verities of crashes that occurred in built-up areas along the rural highways of Nevada. Those factors could aid in making informed decisions while setting up speed zones in these built-up areas. Using descriptive statistics and binary logistic regression model, 337 crashes that occurred in 11 towns along the rural highways from 2002 to 2010 were analyzed. The results showed that more crashes occurred during favorable driving conditions, e.g., 87% crashes on dry roads and 70% crashes in clear weather. The binary logistic regression model showed that crashes occurred from midnight until 4 a.m. were 58.3% likely to be injury crashes rather than property damage only crashes, when other factors were kept at their mean values. Crashes on weekdays were three times more likely to be injury crashes than that occurred on weekends. When other factors were kept at their mean value, crashes involving motorcycles had an 80.2% probability of being injury crashes. Speeding was found to be 17 times more responsible for injury crashes than mechanical defects of the vehicle. As a result of this study, the Nevada Department of Transportation now can take various steps to improve public safety, including steps to reduce speeding and encourage the use of helmets for motorcycle riders.

Effect of time-of-day and day-of-the-week on congestion duration and breakdown:A case study at a bottleneck in Salem,NH

This work uses regression models to analyze two characteristics of recurrent congestion： breakdown, the transition from freely flowing conditions to a congested state, and duration, the time between the onset and clearance of recurrent congestion. First, we apply a binary logistic regression model where a continuous measurement for traffic flow and a dichoto- mous categorical variable for time-of-day （AM- or PM-rush hours） is used to predict the probability of breakdown. Second, we apply an ordinary least squares regression model where categorical variables for time-of-day （AM- or PM-rush hours） and day-of-the-week （Monday-Thursday or Friday） are used to predict recurrent congestion duration. Models are fitted to data collected from a bottleneck on 1-93 in Salem, NH, over a period of 9 months. Results from the breakdown model, predict probabilities of recurrent congestion, are consistent with observed traffic and illustrate an upshift in breakdown probabilities between the AM- and PM-rush periods. Results from the regression model for congestion duration reveal the presences of significant interaction between time-of-day and day-of-the-week. Thus, the effect of time-of-day on congestion duration depends on the day-of-the-week. This work provides a simplification of recurrent congestion and recovery, very noisy processes. Simplification, conveying complex relationships with simple statistical summaries-facts, is a practical and powerful tool for traffic administrators to use in the decision-making process.

Factors affecting farmers’participation in watershed management programs in the Northeastern highlands of Ethiopia:a case study in the Teleyayen sub-watershed

Introduction:This study investigated factors affecting farmers’participation in watershed management programs in the Northeastern highlands of Ethiopia by taking the Teleyayen sub-watershed as a case study.Data were collected from 215 farm households which were selected from the four villages using a multistage sampling procedure,involving a combination of purposive and random sampling.Data were gathered using a structured survey questionnaire,focus group discussion,and key informant interviews.Descriptive analysis,Pearson correlation analysis,and regression analysis were employed to analyze the data.Results:Findings of this study showed that farmer’s perception has a strong positive correlation(r=0.612,P=0.000)with the farmer’s decision to participate in the watershed management programs followed by government support(r=0.163,P=0.017),while the slope of the farmland and the gender of the household head have shown significant and negative associations.The binary logistic regression analysis also revealed that six independent variables were significant in explaining the factors affecting the farmers’decision to participate in watershed management programs.These variables were land redistribution,gender,agricultural labor force,extension service,farm size,and slope.Of these,land redistribution,gender,agricultural labor force,extension service,and slope of the farmland indicated a negative influence,while farm size of a household exerted a positive impact.The study also examined the role of discrete variables in explaining variations of variables in affecting the farmers’decision to participate in the programs.Thus,two variables found to be significant.These variables are the gender of the household head and land tenure security.Accordingly,the chi-square result of the variable(χ^(2)=9.052)of gender was found to be statistically significant at the 95%level of significance.Similarly,the chi-square result(X^(2)=8.792)of land tenure security was found to be statistically significant at the 95%level of significance.Conclusions:The result of the study suggests to work on raising the awareness of farmers’about the long-term benefits of the watershed programs and to design a strategy to diversify their livelihoods.