Applications of Statistics in the Field of General Insurance: An Overview

This is an expository paper on applications of statistics in the field of general insurance, also called non‐life insurance. Unlike life insurance where advanced statistical techniques have long been part of financial mathematics and actuarial applications, their use is only relatively recent in non‐life insurance. The business model of insurance companies, especially those active in non‐life insurance, has seen dramatic changes over the last 15 years. The aim of this paper is to convince the readers that especially today non‐life insurance is not only an exciting ground to apply existing modern statistical tools but also a fertile environment for new and challenging statistical developments. The activities of an insurance company can be viewed as an industrial process where data management and data analysis play a key role. That is why a fundamental understanding of data‐related issues (such as data quality, variability, analysis and correct interpretation) is so essential to the insurance business. These are exactly the tasks where professional statisticians excel. Also, a better understanding of the field of general insurance by statisticians will promote fruitful exchanges between actuaries and statisticians, thereby helping to bring actuarial and statistical professional societies closer to each other. Selected examples are used to cover the essential aspects of general insurance, and all of them are based on the author's experience. The paper concludes with some remarks on the role of statisticians working in general insurance.     

Nearly exact sample size calculation for powerful non‐randomized tests for differences between binomial proportions

In the case of two independent samples, it turns out that among the procedures taken in consideration, BOSCHLOO'S technique of raising the nominal level in the standard conditional test as far as admissible performs best in terms of power against almost all alternatives. The computational burden entailed in exact sample size calculation is comparatively modest for both the uniformly most powerful unbiased randomized and the conservative non‐randomized version of the exact Fisher‐type test. Computing these values yields a pair of bounds enclosing the exact sample size required for the Boschloo test, and it seems reasonable to replace the exact value with the middle of the corresponding interval. Comparisons between these mid‐N estimates and the fully exact sample sizes lead to the conclusion that the extra computational effort required for obtaining the latter is mostly dispensable. This holds also true in the case of paired binary data (McNemar setting). In the latter, the level‐corrected score test turns out to be almost as powerful as the randomized uniformly most powerful unbiased test and should be preferred to the McNemar–Boschloo test. The mid‐N rule provides a fairly tight upper bound to the exact sample size for the score test for paired proportions.     

Selection of influential variables in ordinal data with preponderance of zeros

Presence of excess zero in ordinal data is pervasive in areas like medical and social sciences. Unfortunately, analysis of such kind of data has so far hardly been looked into, perhaps for the reason that the underlying model that fits such data, is not a generalized linear model. Obviously some methodological developments and intensive computations are required. The current investigation is concerned with the selection of variables in such models. In many occasions where the number of predictors is quite large and some of them are not useful, the maximum likelihood approach is not the automatic choice. As, apart from the messy calculations involved, this approach fails to provide efficient estimates of the underlying parameters. The proposed penalized approach includes ?₁ penalty (LASSO) and the mixture of ?₁ and ?₂ penalties (elastic net). We propose a coordinate descent algorithm to fit a wide class of ordinal regression models and select useful variables appearing in both the ordinal regression and the logistic regression based mixing component. A rigorous discussion on the selection of predictors has been made through a simulation study. The proposed method is illustrated by analyzing the severity of driver injury from Michigan upper peninsula road accidents.     

Training and organisational performance: A meta‐analysis of temporal,institutional and organisational context moderators

Drawing on systems theory, we conducted a moderated meta‐analysis of the training and organisational performance relationship using 119 primary studies. We examined the moderating effects of quality versus quantity of training, time, institutional and organisational context factors in the relationship between training and organisational performance. Our findings reveal that training is positively and directly related to organisational performance with no statistically significant difference between measures of training quality and quantity. We found that the relationship was stronger over time and that country performance orientation and country labour cost moderate the training and organisational performance relationship. We found no evidence for the moderating effects of the three organisational context moderators we examined (i.e. industry sector, organisational size and technology intensity). Finally, our results reveal that training type (i.e. general or firm‐specific) does not moderate the training and organisational performance relationship.     

Selecting people based on person‐organisation fit: Implications for intrateam trust and team performance

Teamwork is widely adopted in organisations. Although much evidence indicates that using person‐organisation (P‐O) fit as a selection criterion benefits individual employees, little is known about how this practice influences team functioning. Drawing on the input‐mediator‐outcome model and the research on value congruence, this study built and tested a model that links P‐O fit in recruitment to work teams' performance. Based on data collected from team members, team leaders, human resources managers, and chief executive officers in 96 firms, we found that P‐O fit in recruitment had a positive relationship with team performance and that intrateam trust mediated the relationship between P‐O fit in recruitment and team performance. Further, this mediated relationship existed only when the organisation had a weak, rather than strong, respect‐for‐people culture. This study contributes to the P‐O fit and team literature and has practical implications for human resources practices and team management.     

Corporate financial distress: The case of publicly listed firms in an emerging market economy

Occurrences of financial distress (FD) are not readily obvious yet can span several periods. This paper examines episodes of FD using industry‐relative (IR) firm‐/ accounting‐, market‐ and macro‐level information. Mixed logit regressions reveal that firm‐ and market‐based measures, as well as macro‐level variables explain the likelihood of FD in 263 publicly listed non‐banking firms in the Philippines during the period 1995 to 2018. Rates of identification of firms in financial distressed states of close to 69 percent are obtained at a cutoff probability of 0.30 in the model with time‐varying intercept and slope. This study shows the importance of recognizing heterogeneous firm behavior. The ability to more accurately predict the probability of FD and to determine the financial health of firms can help financial institutions in allocating funds and policy makers in predicting crises episodes.     

Too Little or too much? Exploring U‐shaped Relationships between Corporate Environmental Performance and Corporate Financial Performance

Despite 40 years of research on the relationship between corporate environmental performance (CEP) and corporate financial performance (CFP), there is no generally accepted theoretical framework that explains the contradictory results that have emerged. This unsatisfactory status may be attributed to the fact that linear models dominate the research. Based on an international sample of 2361 firm‐years from 2008 to 2012, we find empirical evidence of a non‐linear, specifically a U‐shaped, relationship between carbon performance and profitability as well as between waste intensity and profitability. The same result holds for the relationship between carbon performance and stock market performance, but solely for manufacturing industries. Our empirical findings provide evidence for the theoretical framework of a ‘too‐little‐of‐a‐good‐thing’ (TLGT) effect, which indicates that the type of relationship (positive, negative) depends on the level of CEP. More precisely, there is a negative CEP–CFP relationship for companies with low CEP and a positive association for high CEP. Copyright © 2015 John Wiley & Sons, Ltd and ERP Environment     

Commitment to Sustainability in Small and Medium‐Sized Enterprises: The Influence of Strategic Orientations and Management Values

Ecosystem degradation and social sustainability have become important issues in the corporate sphere during the last few decades. However, research discussing corporate social responsibility and related concepts has often focused on larger companies, sometimes neglecting the specifics of small and medium‐sized enterprises (SMEs). The main purpose of this study is to examine the relationships between two common strategic orientations, market orientation (MO) and entrepreneurial orientation (EO), in relation to sustainability commitment, sustainability practices and management values in SMEs. Questionnaire responses from 450 Swedish SMEs were analyzed, confirming the influence of MO, EO and sustainability practices on commitment to sustainability, implying that firms committed to sustainability see both market and entrepreneurial advantages of sustainability. The results also show that different parts of MO and EO differ in importance for commitment to sustainability among SMEs. Implications concern the importance for firms and policymakers to work with sustainability issues using both internal and external perspectives. © 2015 The Authors. Business Strategy and the Environment published by ERP Environment and John Wiley & Sons Ltd     

Sustainability Strategy and Eco‐Innovation: A Moderation Model

This study develops a moderation model to examine the role of a proactive environmental strategy on eco‐innovation. Drawing upon the perspectives of contingency theory, this study argues that the impacts of sustainability strategy on eco‐innovation depend on market demand, innovation intensity and government subsidy. The sample used to test the hypotheses is obtained from the Community Innovation Survey in Taiwan. A total of 2955 manufacturing firms are included in the final sample. A logit moderating regression is adopted to analyze the models. The results reveal that market demand and government subsidy positively moderate the relationship between environmental strategy and eco‐innovation. Specifically, firms are more likely to adopt a proactive environmental strategy to improve eco‐innovation under high levels of market demand and government subsidy. Furthermore, the results indicate that innovation intensity affects the effect of environmental strategy on eco‐innovation, but the direction of the influence varies with different categories of eco‐innovation. Copyright © 2016 John Wiley & Sons, Ltd and ERP Environment     

High‐Performance Work Systems in Professional Service Firms: Examining the Practices‐Resources‐Uses‐Performance Linkage

Professional service firms (PSFs) play an important role in the knowledge‐based economy. Their success is highly dependent on their people, the knowledge resources they possess, and how they use these resources. However, how to systematically manage human resources to attain high performance is not fully understood. This study addresses this issue by investigating the linkage mechanisms through which high‐performance work systems (HPWS) influence the performance of PSFs. We integrate resource‐based and dynamic capability theories in order to identify and investigate two intervening mechanisms that link HR practices to firm performance. The first mechanism is the intellectual capital resources comprising the human, social, and organizational capital that HPWS create. The second mechanism is the uses to which both HPWS and resources can be applied, operationalized as organizational ambidexterity, the simultaneous exploitation of existing knowledge and exploration of new knowledge. These mechanisms are hypothesized to link HPWS to firm performance in the form of a practices‐resources‐uses‐performance linkage model. Results from a longitudinal study of 93 accounting firms support this linkage model. © 2015 Wiley Periodicals, Inc.