Advances in Bayesian Time Series Modeling and the Study of Politics: Theory Testing, Forecasting, and Policy Analysis

John R. Freeman Department of Political Science, University of Minnesota, 267 19th Avenue, Minneapolis, MN 55455 e-mail: pbrandt{at}utdallas.edu (corresponding author) e-mail: freeman{at}polisci.umn.edu Bayesian approaches to the study of politics are increasinglypopular. But Bayesian approaches to modeling multiple time serieshave not been critically evaluated. This is in spite of thepotential value of these models in international relations,political economy, and other fields of our discipline. We reviewrecent developments in Bayesian multi-equation time series modelingin theory testing, forecasting, and policy analysis. Methodsfor constructing Bayesian measures of uncertainty of impulseresponses (Bayesian shape error bands) are explained. A referenceprior for these models that has proven useful in short- andmedium-term forecasting in macroeconomics is described. Oncemodified to incorporate our experience analyzing political dataand our theories, this prior can enhance our ability to forecastover the short and medium terms complex political dynamics likethose exhibited by certain international conflicts. In addition,we explain how contingent Bayesian forecasts can be constructed,contingent Bayesian forecasts that embody policy counterfactuals.The value of these new Bayesian methods is illustrated in areanalysis of the Israeli-Palestinian conflict of the 1980s.     

Identification of N,N-dimethylamphetamine formed by methylation of methamphetamine in formalin-fixed liver tissue by multistage mass spectrometry

Methamphetamine is methylated in the presence of unbuffered formalin solutions within hours at room temperature. The product, N,N-dimethylamphetamine, is also found in human liver exposed to methamphetamine followed by incubation with formalin. In the present study, a direct mass spectrometric method was developed to identify N,N-dimethylamphetamine in human liver before and after treatment with formalin. Human liver samples were obtained from four deaths that were investigated by the West Virginia Office of Chief Medical Examiner. Full toxicological analysis was conducted on samples from the decedents and methamphetamine was among the positive findings in each case. The method used to expose liver tissue to formaldehyde involved treating a small piece of liver from each case with formalin solution (20% v/v) for 24 h at room temperature. The formalin treated tissues were homogenized and the resulting suspension was sonicated for 5 min, and then centrifuged. Supernatant aliquots were directly analyzed by electrospray ionization (ESI) mass spectrometry without chromatographic isolation. Positive ion multistage mass spectra recorded in MS, MS/MS and MS/MS/MS (MS3) modes were used to confirm the presence of N,N-dimethylamphetamine and methamphetamine in the mixture. Liver tissue not treated with formalin did not contain a detectable level of N,N-dimethylamphetamine. Decreases in methamphetamine concentrations in liver tissue resulting from treatment with formalin were measured using deuterium-labeled methamphetamine as internal standard. The method can be completed in less than 2 h on thawed tissue. The results suggest that the process of fixing tissues with formalin may lead to false negative findings for methamphetamine.     

Simultaneous analyses of cocaine, cocaethylene, and their possible metabolic and pyrolytic products

A method was developed for simultaneously analyzing cocaine (COC), benzoylecgonine (BZE), norbenzoylecgonine (BNE), norcocaine (NCOC), ecgonine (ECG), ecgonine methyl ester (EME), m-hydroxybenzoylecgonine (HBZE), anhydroecgonine methyl ester (AEME), cocaethylene (CE), norcocaethylene (NCE), and ecgonine ethyl ester (EEE) in blood, urine, and muscle. Available deuterated analogs of these analytes were used as internal standards. Proteins from blood and muscle homogenate were precipitated with cold acetonitrile. After the removal of acetonitrile by evaporation, the supernatants and urine were subjected to solid-phase extraction. The eluted analytes were converted to their hydrochloride salts and derivatized with pentafluoropropionic anhydride and 2,2,3,3,3-pentafluoro-1-propanol. The derivatized products were analyzed by a gas chromatograph (GC)/mass spectrometer by selected ion monitoring. The limit of detection (LOD) for COC, BZE, NCOC, EME, CE, NCE, and EEE was 2ng/ml, while the LODs for BNE, ECG, HBZE, and AEME were 25, 640, 50, and 13 ng/ml, respectively. This method was successfully applied in analyzing 13 case samples from aviation accident pilot fatalities and motor vehicle operators. AEME concentrations found in the 13 samples were consistent with those produced solely by the GC inlet pyrolysis of COC controls in blood. Anhydroecgonine cannot be used as a marker for the abuse of COC by smoking because it is also pyrolytically produced from COC metabolites on the GC inlet. The developed method can be effectively adopted for analyzing COC and related compounds in urine, blood, and muscle by a single extraction with increased sensitivity through formation of hydrochloride salts and using a one-step derivatization.     

Party composition in the American states: Clarifying concepts and explaining changes in partisanship since the 1950s

Students of state politics have long been interested in the partisanship and the degree of interparty competition in the American states. This has led to numerous attempts to quantify state-level partisanship and competition, the most significant by Ranney (1965). Interestingly, however, scholars have never specified clearly the relationship between these two concepts as measured by the Ranney index. This research attempts to do just that and reveals that the two are different measures which are related systematically. More important, understanding changes in a state's partisanship is determined to be crucial to understanding changes in its level of two-party competition. As such, in order to explain changes in partisanship and competition across the American states from the 1950s to the 1970s, we focus on accounting for changes in partisanship. To this end, changes in aggregate demographic variables account very well for states' movement along the Ranney index over the last 3 decades.The names of the authors appear in alphabetical order and imply that this paper is in every way a collaborative enterprise.     

Can ministries of health support primary health care? Some suggestions for structural reorganization and planning

Patrick Vaughan is Reader in health care epidemiology, Gill Walt is Lecturer in health policy and Anne Mills is Lecturer in health economics, at the Evaluation and Planning Centre for Health Care, London School of Hygiene and Tropical Medicine, Gower Street, London WC1, United Kingdom. A previous version of this paper was prepared for the Commonwealth Secretariat, London, for the Conference of Commonwealth Health Ministers held in Ottawa, Canada, in October 1983.     

Regulators and Redskins

We examine the correlation between federal government activity and performance of the capital’s National Football League team, the Washington Redskins. We find a positive, non-spurious, and robust correlation between the Redskins’ winning percentage and bureaucratic output, measured by pages published in the Federal Register. Because the Redskins’ performance is prototypically exogenous, we give this result a causal interpretation and provide a plausible, causal mechanism: bureaucrats must make “logrolling” deals to expand their regulatory power, and a winning football team offers a shared source of optimism to lubricate such negotiations. We do not find the same correlation when examining congressional activity.