Marketing managers and their companies are better served by spending less on building brand loyalty up front and maintaining a reserve for advertising during a post-crisis period. Further, ad spending after a crisis is more effective in building brand interest than before a crisis.
What is the optimal advertising budget and allocation that maximizes profits across multiple regions and over time? The chief marketing officer of a Fortune 500 company raised the question after she noticed that increasing her company’s advertising expenditures enhanced sales as expected, while profits diminished.
Can using common accounting practices, like gathering data and using performance metrics, result in improved student performance?
Professor Shannon Anderson is exploring this question as part of a research project examining the use of a data portal tool created by Aspire Public Schools. The system, known as Schoolzilla, collects raw data from a school’s source systems—including the California State Test and benchmark assessments—and makes it readily available to teachers and administrators in the form of charts, graphs and textual information. The idea is to empower teachers to immediately identify which students are doing well, which are struggling, and to adjust their teaching strategies. The data also will make it possible to evaluate individual or groups of students over the long term.
India’s rapidly expanding middle class and steady increase in household disposable income has attracted the interest of multinational retailers like IKEA, Apple, WalMart and Tesco. But estimating consumer demand has been challenging, says Professor Prasad Naik, and the standard methods being used have proved unreliable.
In this paper, we propose two important measures, quantile correlation (QCOR) and quantile partial correlation (QPCOR). We then apply them to quantile au- toregressive (QAR) models, and introduce two valuable quantities, the quantile autocorrelation function (QACF) and the quantile partial autocorrelation function (QPACF). This allows us to extend the Box-Jenkins three-stage procedure (model identification, model parameter estimation, and model diagnostic checking) from classical autoregressive models to quantile autoregressive models.
A Bayesian Information Criterion for Portfolio Selection
Computational Statistics & Data Analysis, 2012
The 1952 mean–variance theory of Markowitz indicates that large investment portfolios naturally provide better risk diversification than small ones. However, due to parameter estimation errors, one may find ambiguous results in practice. Hence, it is essential to identify relevant stocks to alleviate the impact of estimation error in portfolio selection.
Profiled Forward Regression for Ultra-High Dimensional Variable Screening in Semiparametric Partially Linear Models
Statistica Sinica, 2012
In partially linear model selection, Professor Chih-Ling Tsai and co-authors Hua Liang from the University of Rochester and Hansheng Wang from Peking University develop a profiled forward regression (PFR) algorithm for ultrahigh dimensional variable screening. The PFR algorithm effectively combines the ideas of nonparametric profiling and forward regression.
In multivariate analysis, the covariance matrix associated with a set of vari- ables of interest (namely response variables) commonly contains valuable infor- mation about the dataset. When the dimension of response variables is con- siderably larger than the sample size, it is a non-trivial task to assess whether they are linear relationships between the variables. It is even more challenging to determine whether a set of explanatory variables can explain those relation- ships.
By Tove H. Hammer, Steven C. Currall & Robert N. Stern