Inference with Dependent Data in Accounting and Finance Applications

We review developments in conducting inference for model parameters in the presence of intertemporal and cross‐sectional dependence with an emphasis on panel data applications. We review the use of heteroskedasticity and autocorrelation consistent (HAC) standard error estimators, which include the standard clustered and multiway clustered estimators, and discuss alternative sample‐splitting inference procedures, such as the Fama–Macbeth procedure, within this context. We outline pros and cons of the different procedures. We then illustrate the properties of the discussed procedures within a simulation experiment designed to mimic the type of firm‐level panel data that might be encountered in accounting and finance applications. Our conclusion, based on theoretical properties and simulation performance, is that sample‐splitting procedures with suitably chosen splits are the most likely to deliver robust inferential statements with approximately correct coverage properties in the types of large, heterogeneous panels many researchers are likely to face.     

Bank net interest margins,the yield curve,and the 2007–2009 financial crisis

Using quarterly call report data from 2000 to 2016, we reexamine the relationship between net interest margins (NIM) and the yield curve for more than 5,500 U.S. commercial banks. In the full sample, yield curve and RGDP growth have positive effects on NIM, while inflation and deposit‐to‐loan ratios (D/L) have negative effects. Splitting the sample around the 2007–2009 crisis, we show the impact of yield curve and RGDP growth on NIM increasing during the “recovery” (2009Q3 to 2016Q4), and inflation and D/L changing signs. Positive effects of yield curve on profits vary with bank size and change over time.     

A logistic regression point of view toward loss given default distribution estimation

We propose a new procedure to estimate the loss given default (LGD) distribution. Owing to the complicated shape of the LGD distribution, using a smooth density function as a driver to estimate it may result in a decline in model fit. To overcome this problem, we first apply the logistic regression to estimate the LGD cumulative distribution function. Then, we convert the result into the LGD distribution estimate. To implement the newly proposed estimation procedure, we collect a sample of 5269 defaulted debts from Moody’s Default and Recovery Database. A performance study is performed using 2000 pairs of in-sample and out-of-sample data-sets with different sizes that are randomly selected from the entire sample. Our results show that the newly proposed procedure has better and more robust performance than its alternatives, in the sense of yielding more accurate in-sample and out-of-sample LGD distribution estimates. Thus, it is useful for studying the LGD distribution.     

Properties of Census Dual System Population Size Estimators

We study parametric and non‐parametric approaches for assessing the accuracy and coverage of a population census based on dual system surveys. The two parametric approaches being considered are post‐stratification and logistic regression, which have been or will be implemented for the US Census dual system surveys. We show that the parametric model‐based approaches are generally biased unless the model is correctly specified. We then study a local post‐stratification approach based on a non‐parametric kernel estimate of the Census enumeration functions. We illustrate that the non‐parametric approach avoids the risk of model mis‐specification and is consistent under relatively weak conditions. The performances of these estimators are evaluated numerically via simulation studies and an empirical analysis based on the 2000 US Census post‐enumeration survey data.     

The moments of the Gompertz distribution and maximum likelihood estimation of its parameters

The Gompertz distribution is widely used to describe the distribution of adult deaths. Previous works concentrated on formulating approximate relationships to characterise it. However, using the generalised integro-exponential function, exact formulas can be derived for its moment-generating function and central moments. Based on the exact central moments, higher accuracy approximations can be defined for them. In demographic or actuarial applications, maximum likelihood estimation is often used to determine the parameters of the Gompertz distribution. By solving the maximum likelihood estimates analytically, the dimension of the optimisation problem can be reduced to one both in the case of discrete and continuous data. Monte Carlo experiments show that by ML estimation, higher accuracy estimates can be acquired than by the method of moments.     

Economic contribution of essential air service flights on small and remote communities

The Essential Air Service Program (EAS) has attracted considerable criticism and has been a target for either modification or complete termination almost since its inception through the Airline Deregulation Act in 1978. Although its opponents emphasize the program's inefficiency, its supporters claim that the program is crucial to accessing small and remote communities, which helps them develop economically and socially. This paper demonstrates the economic contributions of EAS flights to small and remote communities. Using a two-stage least squares estimation, the major findings indicate that a 1% increase in air passenger traffic in EAS airports with a minimum annual air passenger traffic of 1000 likely leads to a 0.12% increase in per capita income of the community served by that airport. Our results also suggest that EAS communities that are able to sustain their subsidized flights experienced higher per capita income growth in the 1999–2011 period than did ex-EAS communities that lost their flights as a result of non-eligibility.     

The impact of population growth and climate change on food security in Africa: looking ahead to 2050

Providing nutritious and environmentally sustainable food to all people at all times is one of the greatest challenges currently facing society. This problem is particularly acute in Africa where an estimated one in four people still lack adequate food to sustain an active and healthy life. In this study, we consider the potential impact of future population growth and climate change on food security in Africa, looking ahead to 2050. A modelling framework termed FEEDME (Food Estimation and Export for Diet and Malnutrition Evaluation) was used which was characterized to model the impacts of future climate changes (utilizing the Intergovernmental Panel on Climate Change Special Report on Emissions Scenarios projections) and projected population growth on food availability and subsequent undernourishment prevalence in 44 African countries. Our results indicate that projected rapid population growth will be the leading cause of food insecurity and widespread undernourishment across Africa. Very little to no difference in undernourishment projections were found when we examined future scenarios with and without the effects of climate change, suggesting population growth is the dominant driver of change. Various adaptation options are discussed, such as closing the yield gap via sustainable intensification and increasing imports through trade and aid agreements. These strategies are likely to be critical in preventing catastrophic future food insecurity.     

Forecast combinations under structural break uncertainty

This paper proposes two new weighting schemes that average forecasts based on different estimation windows in order to account for possible structural change. The first scheme weights the forecasts according to the values of reversed ordered CUSUM (ROC) test statistics, while the second weighting method simply assigns heavier weights to forecasts that use more recent information. Simulation results show that, when structural breaks are present, forecasts based on the first weighting scheme outperform those based on a procedure that simply uses ROC tests to choose and forecast from a single post-break estimation window. Combination forecasts based on our second weighting scheme outperform equally weighted combination forecasts. An empirical application based on a NAIRU Phillips curve model for the G7 countries illustrates these findings, and also shows that combination forecasts can outperform the random walk forecasting model.     

Does the labor-income process contain a unit root? Evidence from individual-specific time series

Calibrations of models related to life-cycle behavior of consumption and saving often invoke the important assumption of a unit root in individuals׳ labor-income process. We for the first time test this assumption using methods for univariate time series. Based on longitudinal register data from 1968 to 2005, we first estimate an autoregressive model for each individual using a method for approximately median-unbiased estimation. We then exploit the resulting distribution of the individual-specific estimates to draw inference about the presence of a unit root. Results indicate that earnings for the representative worker are governed by a process where shocks to earnings have moderate persistence and are both economically and statistically significantly different from having permanent effects. These results question the heavy use of unit-root processes for earnings.     

Public sector enterprise disinvestment in India: Efficiency gains in a political context

India instituted a program of state enterprise disinvestment in 1991 as part of a sweeping reform initiative. This study analyses the effect of disinvestment on enterprise performance, conditioned on political context as characterized by the ideological leanings of the parties in power at state and central levels. Using stochastic frontier analysis, measures of firm efficiency are generated for 238 central public sector enterprises for the period 1991–1992 to 2010–2011. The relationship between efficiency so measured and disinvestment is then estimated using a two-stage instrumental variables approach to control for endogeneity between firm efficiency and selection for disinvestment. Initial disinvestment is associated with substantial efficiency gains, but subsequent disinvestment much less so and the proportion of shares disinvested only loosely so. This may be explained by the transformative effect of initial stock market listing on accountability and profit orientation that is not repeated with further rounds of disinvestment. The effect of disinvestment on performance is stronger if the enterprise is located in a state governed by a right leaning party or one that is ideologically aligned with the party in power at the centre.