Considerations for using multiple imputation in propensity score-weighted analysis

We present our considerations for using multiple imputation to account for missing data in propensity score-weighted analysis with bootstrap percentile confidence interval. We outline the assumptions underlying each of the methods and discuss the methodological and practical implications of our choices and briefly point to alternatives. We made a number of choices a priori for example to use logistic regression-based propensity scores to produce standardized mortality ratio-weights and Substantive Model Compatible-Full Conditional Specification to multiply impute missing data (given no violation of underlying assumptions). We present a methodology to combine these methods by choosing the propensity score model based on covariate balance, using this model as the substantive model in the multiple imputation, producing and averaging the point estimates from each multiple imputed data set to give the estimate of association and computing the percentile confidence interval by bootstrapping. The described methodology is demanding in both work-load and in computational time, however, we do not consider the prior a draw-back: it makes some of the underlying assumptions explicit and the latter may be a nuisance that will diminish with faster computers and better implementations.

RESEARCH OF THE DOWNTIME OF PUBLIC TRANSPORT VEHICLES AT STOP IN LVIV

Understanding the features and availability of information about the functioning indicators of public transport stops will improve the organization of movement in the city. This paper intended to study the trends change the downtime of public transport vehicles at stop using data of field research. The analysis of literary sources has shown that the number of studies of combined stops is insufficient and we tried to concentrate on the comparison of bus and combined stops. The field inspections were conducted at the city’s stops in Lviv on working days separately in the morning peak time and separately in the interpeak dinner period. For the given series of data of field inspections the following statistical characteristics were calculated: average value, median, percentile, confidence interval, value asymmetry. On the basis of research, it was discovered that other factors, other than the time of boarding and landing of passengers, affect the duration of downtime. And the influence of these factors is more than half the duration of the entire downtime. It was found that the downtimes are longer for combined bus and tram stops. When constructing trend lines to determine the relationship between passenger traffic and the downtime of public transport vehicles at stop the value of the determination factor in all cases does not exceed 0.52. Dependence on the sufficient value of the determination factor is found for the interconnection between the average downtime and number of vehicles arrivals per hour and between the average downtime and the stop passenger traffic per hour. The results obtained should be taken into account when calculating the capacity of public transport stops, lines of public transport routes, links of the street and road network. Keywords: downtime, stop of public transport, passenger exchange, median.