Joint modeling of longitudinal continuous,longitudinal ordinal,and time-to-event outcomes

Lifetime Data Analysis - In this paper, we propose an innovative method for jointly analyzing survival data and longitudinally measured continuous and ordinal data. We use a random effects...     

Expert Testimony Influences Juror Decisions in Criminal Trials Involving Recovered Memories of Childhood Sexual Abuse

We examined the impact of expert witness orientation (researcher or clinical practitioner) and type of testimony (testimony for the prosecution, defense, both prosecution and defense, or no testimony) on mock jurors’ decisions in a sexual abuse trial. Participants acted as mock jurors on a sexual abuse criminal trial based on recovered memory that included expert witness testimony. Results showed that expert witness testimony provided by a researcher did not impact mock jurors’ guilt ratings any differently than the expert witness testimony provided by a clinical practitioner. However, type of testimony had a significant effect on jurors’ guilt ratings such that jurors who read only defense or only prosecution testimony made decisions favoring the relevant side.     

Inflation in Transition Economies: An Empirical Analysis

The present research works with yearly data on backward looking measures of inflation i.e. GDP deflator series for Bulgaria, Czech Republic, Estonia, Hungary, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, and Slovakia. Because of the data limitations in the transition economies annual series are converted to smooth quarterly series before they are employed for inflation prognosis using state space models. The results show statistically significant evidence of time-varying volatility in Bulgaria, Czech Republic, Estonia, Hungary, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, and Slovakia using critical values obtained from χ₃² distributions. The results also show that the non-normality hypothesis cannot be rejected in all the series using critical values due McCulloch (J Bus Econ Stat 15:74–81, 1997). The efficiently estimated inflation forecast range between 0.925 for Bulgaria to 99.106 for Latvia. The estimated value of the characteristic exponent α of 2.000 for Hungary, and for the other countries having α equal to 1.999 shows tendency toward normal behavior in exception of Bulgaria, where α equals 1.861.      

Stock Returns Predictability in Transition Economies

I investigate the possible existence of predictable components in Croatia, Hungary, Latvia, Russia, Slovakia, and Ukraine excess returns over the relevant risk-free rates using non-Gaussian state space or unobservable-component models with stable distributions and volatility persistence. The results show that most stock markets in these countries encompass volatility persistence. Moreover, these markets have a stable characteristic exponent α ranging from 1.607 for Hungary to 1.749 for Latvia showing heavy tails. However, for Russia the characteristic exponent α of 1.999 is close to the value pertaining to normal behavior. Our findings reveal that even after accounting for nonnormality and volatility persistence, a predictable signal in return exists in Croatia, Hungary, Latvia, Slovakia, and Ukraine at the 5 percent level of significance using critical values from Monte Carlo simulations. The efficient estimated excess returns range from 0.003 (0.036) percent per month (annum) for Ukraine to 0.094 (1.128) percent per month (annum) for Russia.     

Testing equality of two beta binomial proportions in the presence of unequal extra-dispersion parameters

Data in the form of proportions with extra-dispersion (over/under) arise in many biomedical, epidemiological, and toxicological applications. In some situations, two samples of data in the form of proportions with extra-dispersion arise in which the problem is to test the equality of the proportions in the two groups with unspecified and possibly unequal extra-dispersion parameters. This problem is analogous to the traditional Behrens-Fisher problem in which two normal population means with possibly unequal variances are compared. To deal with this problem we develop eight tests and compare them in terms of empirical size and power, using a simulation study. Simulations show that a C(α) test based on extended quasi-likelihood estimates of the nuisance parameters holds nominal level most effectively (close to the nominal level) and it is at least as powerful as any other statistic that is not liberal. It has the simplest formula, is based on estimates of the nuisance parameters only under the null hypothesis, and is easiest to calculate. Also, it is robust in the sense that no distributional assumption is required to develop this statistic.