Statistical Models Commons^™

A Modified Version Of The Lewellen And Badrinath Measure Of Tobin's Q, Darrell Lee, James Tompkins

James Tompkins

Lewellen and Badrinath (1997) propose a superior method of measuring Tobin's Q. Unfortunately, their method is prone to a high percentage of missing observations and results in selecting samples of larger and more mature firms with lower Q statistics. A slight modification is proposed that preserves the appeal of their method, yet almost doubles the sample size, avoids sampling problems, and is statistically indistinguishable from their Q measure. In addition, a step in the Lewellen and Badrinath Q calculation is clarified, which was inadvertently omitted in their explanation, and, if left undone, can result in downward-biased measures of Q.

Identifying Key Variables And Interactions In Statistical Models Of Building Energy Consumption Using Regularization, David Hsu

David Hsu

Statistical models can only be as good as the data put into them. Data about energy consumption continues to grow, particularly its non-technical aspects, but these variables are often interpreted differently among disciplines, datasets, and contexts. Selecting key variables and interactions is therefore an important step in achieving more accurate predictions, better interpretation, and identification of key subgroups for further analysis.

This paper therefore makes two main contributions to the modeling and analysis of energy consumption of buildings. First, it introduces regularization, also known as penalized regression, for principled selection of variables and interactions. Second, this approach is demonstrated by …

A Comparative Analysis Of Decision Trees Vis-À-Vis Other Computational Data Mining Techniques In Automotive Insurance Fraud Detection, Adrian Gepp, Kuldeep Kumar, J Holton Wilson, Sukanto Bhattacharya

Kuldeep Kumar

No abstract provided.

A Probabilistic Predictive Model For Residential Mobility In Australia, Mohammad-Reza Namazi-Rad, Nagesh Shukla, Albert Munoz, Payam Mokhtarian, Jun Ma

Payam Mokhtarian

Household relocation modelling is an integral part of the planning process as residential movements influence the demand for community facilities and services. Department of Families, Housing, Community Services and Indigenous Affairs (FaHCSIA) created the Household, Income and Labour Dynamics in Australia (HILDA) program to collect reliable longitudinal data on family and household dynamics. Socio-demographic information (such as general health situation and well-being, lifestyle changes, residential mobility, income and welfare dynamics, and labour market dynamics) is collected from the sampled individuals and households. The data shows that approximately 17% of Australian households and 13% of couple families in the HILDA sample …

Approximate Bayesian Computation In State Space Models, Gael Martin, Brendan Mccabe, Christian Robert, Worapree Ole Maneesoonthorn

Worapree Ole Maneesoonthorn

A new approach to inference in state space models is proposed, based on approximate Bayesian computation (ABC). ABC avoids evaluation of the likelihood function by matching observed summary statistics with statistics computed from data simulated from the true process; exact inference being feasible only if the statistics are sufficient. With finite sample sufficiency unattainable in the state space setting, we seek asymptotic sufficiency via the maximum likelihood estimator (MLE) of the parameters of an auxiliary model. We prove that this auxiliary model-based approach achieves Bayesian consistency, and that - in a precise limiting sense - the proximity to (asymptotic) sufficiency …

Time Series, Unit Roots, And Cointegration: An Introduction, Lonnie K. Stevans

Lonnie K. Stevans

The econometric literature on unit roots took off after the publication of the paper by Nelson and Plosser (1982) that argued that most macroeconomic series have unit roots and that this is important for the analysis of macroeconomic policy. Yule (1926) suggested that regressions based on trending time series data can be spurious. This problem of spurious correlation was further pursued by Granger and Newbold (1974) and this also led to the development of the concept of cointegration (lack of cointegration implies spurious regression). The pathbreaking paper by Granger (1981), first presented at a conference at the University of Florida …

A Comparative Analysis Of Decision Trees Vis-À-Vis Other Computational Data Mining Techniques In Automotive Insurance Fraud Detection, Adrian Gepp, Kuldeep Kumar, J Holton Wilson, Sukanto Bhattacharya

Adrian Gepp

No abstract provided.