On the natural selection rule in general linear models

We consider a problem of selecting the best treatment in a general linear model. We look at the properties of the natural selection rule. It is shown that the natural selection rule is minimax under to “0–1” loss function and it is a Bayes rule under a monotone permutation invariant loss function with respect to a permutation invariant prior for every variance balanced design. Some other condition on the design matrix is given so that a Bayes rule with respect to a normal prior will be of simple structure.     

On some noncentral distribution problems for the mixture of two normal populations

Summary The null density of the multiple correlation coefficient when sampling from a mixture of two normal populations has been derived by Srivastava (1983). However, he does not express this density in a rather standard form. The present paper obtains the nonnull densities of the multiple correlation coefficient and the partial correlation coefficient in a rather standard form when sampling from a mixture of two normal populations.     

A product Pareto distribution

The Pareto distributions are becoming increasing prominent in several applied areas. In this note, a new Pareto distribution is introduced. It takes the form of the product of two Pareto probability density functions. Various structural properties of this distribution are derived, including its cumulative distribution function, moments, mean deviation about the mean, mean deviation about the median, entropy, asymptotic distribution of the extreme order statistics, method of moments estimates, maximum likelihood estimates and the Fisher information matrix. The calculations involve the use of several special functions.     

Costly Tax Enforcement and Financial Repression

Using an overlapping generations production‐economy model characterized by financial repression, purposeful government expenditures and cost of tax collection, we analyse whether financial repression can be explained by the cost of raising taxes. We show that with public expenditures affecting utility of the agents, modest costs of tax collection tend to result in financial repression being pursued as an optimal policy by the consolidated government. However, when public expenditures are purposeless, the above result only holds for relatively higher costs of tax collection. But, more importantly, costs of tax collection cannot produce a monotonic increase in the reserve requirements. What are critical, in this regard, are the weights the consumer assigns to the public good in the utility function and the size of the government.     

Are Donor Countries Giving More or Less Aid?

The volume of foreign aid has increased during the last four decades, albeit with interruptions in certain years. Over time, the major recipients have changed: while the share of aid to Asia has diminished since the 1980s, that destined for sub‐Saharan Africa has grown. There is some evidence that, since the late 1990s, debt relief has assumed a larger share of the increased aid flows to sub‐Saharan Africa. The share of technical cooperation—a component of aid that is viewed as being driven by donors—has risen. More recently, there has been an increased emphasis on providing budget support to recipient governments, especially in the form of debt relief. Donor harmonization, national ownership of development plans, and sound policies on the part of the recipients are crucial for the aid to be effective in reducing poverty.     

A cross-country comparison of corporate governance and firm performance: Do financial structure and the legal system matter?

We blend the corporate governance and the financial structure/legal system literature streams to study whether firm performance is enhanced when its governance structure embodies the demands of the host country’s financial structure and legal system. Using a sample of 1736 unique firms representing 22 countries, we find that the joint effect of a country’s financial structure and legal system does matter when explaining the relationship between performance and the overall level of corporate governance in a given country. The results also suggest that firms operating in the market/common combination countries tend to command higher market valuations than firms with a comparable level of corporate governance that operate in the bank/civil combination countries.     

A large factor model for forecasting macroeconomic variables in South Africa

This paper uses large Factor Models (FMs), which accommodate a large cross-section of macroeconomic time series for forecasting the per capita growth rate, inflation, and the nominal short-term interest rate for the South African economy. The FMs used in this study contain 267 quarterly series observed over the period 1980Q1-2006Q4. The results, based on the RMSEs of one- to four-quarter-ahead out-of-sample forecasts from 2001Q1 to 2006Q4, indicate that the FMs tend to outperform alternative models such as an unrestricted VAR, Bayesian VARs (BVARs) and a typical New Keynesian Dynamic Stochastic General Equilibrium (NKDSGE) model in forecasting the three variables under consideration, hence indicating the blessings of dimensionality.     

Tax evasion and financial repression

Using a simple overlapping generations framework, calibrated to four Southern European countries, we analyze the relationship between tax evasion, determined endogenously, and financial repression. We show that higher degree of tax evasion within a country, resulting from a higher level of corruption and a lower penalty rate, yields higher degrees of financial repression as a social optimum. However, a higher degree of tax evasion, due to a lower tax rate, reduces the severity of the financial restriction.     

Leveraging Information Across Categories

Companies are collecting increasing amounts of information about their customers. This effort is based on the assumption that more information is better and that this information can be leveraged to predict customers' behavior in a variety of situations and product categories. For example, information about a customer's purchase behavior in one category can be helpful in predicting his potential behavior in a related category, which in turn could help a firm in its cross-selling efforts. In this paper, we present a model to better understand and predict a consumer's purchases and preferences when we may have limited or no information about him in one or more product categories. Conceptually this involves leveraging information from purchases of other consumers in multiple categories as well as partial information (e.g., purchase in one of the categories) of the target consumer. Our approach builds on the pioneering work of (Rossi et al. (1996)) who demonstrate the value of purchase information in the context of a single product category. We present results from an extensive simulation as well as an application on scanner panel data. Our simulation shows many interesting and somewhat surprising results. Specifically, we find that compared to a single-category analysis, a cross-category analysis does not lead to any significant improvement in data likelihood in most cases. Therefore, the single-category analysis of (Rossi et al. (1996)) is even more powerful than previously thought. However, we also find that a cross-category analysis does improve parameter recovery in many situations as compared to a single-category analysis. It is in these conditions that retailers can use cross-category information to better implement micro marketing programs. We demonstrate the transfer of information across categories in an application of two grocery products—Breakfast Foods and Table Syrup. In spite of a reasonable correlation (0.21) in the price parameter across these two categories, our simulation guidelines predict very little benefit of cross-category analysis over single-category analysis. Our empirical results confirm this prediction.     

Misspecified polynomial detrending

This letter shows that substantial errors result in linear detrending of the data when a linear trend is inappropriate. These errors increase without bound with increasing sample size! Also, the reliability of the linear trend is overstated.