More Offenders, More Crime: Estimating the Size of the Offender Population in a Latin American Setting

Research that has examined the high levels of crime experienced in Latin American settings has suggested that macrostructural variables (such as social inequality), and factors associated with development and institutional capacity, offer explanations for these high crime levels. Although useful, these studies have yet to quantify how these explanations translate to the dynamics of offending activities. In the current study, we examine a key component related to offending dynamics: the size of the offender population. Using two capture-recapture techniques and a bootstrap simulation, estimates were generated of the sizes of the offender populations for three comparable cities in Brazil, Mexico, and England. Each of the techniques generated similar estimates for the offender population size for each city, but with these estimates varying substantially between the cities. This included the estimated offender population size for the Brazilian city being twenty-five times greater than that for the English city. Risk of arrest values were also generated, with these calculated to be substantially lower for the Brazilian and Mexican cities than for the English city. The results provide a quantification of criminal behavior that offers a potential new insight into the high levels of crime that are experienced in Latin American settings.

Cutoff points of T1 rho / T2 mapping relaxation times distinguishing early-stage and advanced osteoarthritis

IntroductionThe histopathology grading system is the gold post-operative method to evaluate cartilage degeneration in knee osteoarthritis (OA). Magnetic resonance imaging (MRI) T1 rho/T2 mapping imaging can be used as a preoperative detection. An association between histopathology and T1 rho/T2 mapping relaxation times value was suggested in previous research. However, the cutoff point was not determined among different histopathology grades. Our study was to discuss the cutoff point of T1 rho/T2 mapping.Material and methodsT1 rho/T2 mapping images were acquired from 80 samples before total knee replacements. Then the histopathology grading system was applied.ResultsThe mean T1 rho/T2 mapping relaxation times of 80 samples were 39.17 ms and 37.98 ms respectively. Significant differences were found in T1 rho/T2 mapping values between early-stage and advanced OA (P < 0.001). The cutoff point for T1 rho was at 33 ms with a sensitivity of 94.12 (95%CI: 80.3 to 99.3) and a specificity of 91.30 (95%CI: 79.2 to 97.6). The cutoff point for T2 mapping was suggested at 35.04 ms with a sensitivity of 88.24 (95%CI: 72.5 to 96.7) and specificity of 97.83 (95%CI: 88.5 to 99.9). After bootstrap simulation, 95% CI of T1 rho/T2 mapping cutoff point was estimated as 29.36 to 36.32 ms and 34.8 to 35.04 ms respectively. The area under PR curve of T1 rho/T2 mapping was 0.972 (95%CI: 0.925 to 0.992) and 0.949 (95%CI: 0.877 to 0.989) respectively.ConclusionsThe cutoff point of T1 rho relaxation times, which was suggested as 33 ms could be used to distinguish early-stage and advanced OA.

A New Intuition into Tourism-Inclusive Growth Nexus in Turkey and Nigeria (1995 – 2018)

This paper examines the symmetric and asymmetric causal relationships between tourism and inclusive growth in Turkey and Nigeria over the period 1995Q1-2018Q4. The study employs a bootstrap simulation method with leverage adjustments to achieve the objective of the study. The method is used to see whether positive or negative tourism shocks cause inclusive growth and whether positive or negative inclusive growth shocks cause tourism activity. The results show no evidence of asymmetric causality between tourism and inclusive growth, while there is evidence of symmetric causality running from tourism to inclusive growth in Turkey. On the other hand, there is neither symmetric nor asymmetric causal relationship between tourism and inclusive growth in Nigeria. In sum, both neutrality and tourism-led growth hypothesis hold in Turkey, while Nigeria gives credence to neutrality hypothesis. The recommendations coming from the findings are that the tourism sector in both countries, Nigeria in particular, should be repositioned for better performance and effectiveness in stimulating inclusive growth. Rather than focusing on pro-poor and micro-based tourism policies that favour selected communities and localities, tourism should be included in development plans nationally, in order to ensure wider participation and more encompassing trickle-down effects on the citizenry. Furthermore, both countries should implement policies that will stimulate their tourism sectors for a larger and more significant contribution to real GDP.

Lymph node metastases of gastric cancer and blood cell circuit.

e16024 Background: Significance of blood cell circuit in terms of detection of gastric cancer (GC) patients (GCP) with lymph node metastases was investigated. Methods: We analyzed data of 793 consecutive GCP (age = 57±9.4 years; tumor size = 5.4±3.1 cm) radically operated (R0) and monitored in 1975-2021 (m = 555, f = 238; distal gastrectomies = 460, proximal gastrectomies = 163, total gastrectomies = 170, combined gastrectomies with resection of pancreas, liver, diaphragm, colon transversum, esophagus, duodenum, splenectomy, small intestine, kidney, adrenal gland = 244; D2-lymphadenectomy = 513, D3-4 = 280; T1 = 235, T2 = 220, T3 = 182, T4 = 156; N0 = 433, N1 = 109, N2 = 251; G1 = 222, G2 = 162, G3 = 409; early GC = 162, invasive = 631; only surgery = 621, adjuvant chemoimmunotherapy-AT = 172 (5-FU+thymalin/taktivin). Variables selected for study were input levels of blood cell circuit, sex, age, TNMG. Differences between groups were evaluated using discriminant analysis, clustering, nonlinear estimation, structural equation modeling, Monte Carlo, bootstrap simulation and neural networks computing. Results: It was revealed that separation of GCP with lymph node metastases (n = 360) from GCP without metastases (n = 433) significantly depended on: eosinophils (%, abs, total), thrombocytes (abs, total), ESS, Hb, erythrocytes (abs), residual nitrogen, protein, cell ratio factors (CRF) (ratio between cancer cells- CC and blood cells subpopulations), T, G, tumor size, histology, tumor growth, blood group, procedures type (P = 0.043-0.000). Neural networks computing, genetic algorithm selection and bootstrap simulation revealed relationships of lymph node metastases and CRF: thrombocytes/CC (rank = 1), healthy cells/CC (2), erythrocytes/CC (3), monocytes/CC (4), segmented neutrophils/CC (5), lymphocytes/CC (6), leucocytes/CC (7), eosinophils/CC (8), stick neutrophils/CC (9). Correct classification N0—N12 was 99.9% by neural networks computing (area under ROC curve = 1.0; error = 0.0). Conclusions: Lymph node metastases of gastric cancer significantly depended on blood cell circuit.

Firm Characteristics and Empirical Factor Models: A Model Mining Experiment

In a novel model mining experiment, we data mine hundreds of randomly constructed three-factor models and find that many outperform well-known models from the literature, including those with four and five factors. The results provide compelling evidence that the threshold of factor model success needs to be raised. Confidence intervals for model rankings, derived from a bootstrap simulation, offer new insights into the consistency of a model’s pricing ability. Rankings for some well-known models are unusually volatile, which have wider confidence intervals than that of most of the random factor models.

Precise early detection of gastric cancer and blood cell circuit.

e16512 Background: Significance of blood cell circuit in terms of early detection of gastric cancer (GC) was investigated. Methods: In trial (1975-2020) consecutive cases after surgery, monitored 136 GC patients (GCP) (m = 90, f = 46; distal gastrectomies = 95, proximal gastrectomies = 34, total gastrectomies = 7) with pathologic stage IA (tumor size = 1.81±0.70 cm; adenocarcinoma = 136; T1N0M0 = 136; G1 = 67, G2 = 26, G3 = 43, 5-year survival = 100%) and 120 healthy donors (HD) (m = 69, f = 51) were reviewed. Variables selected for study were input levels of blood cell circuit, sex, age, TNMG. Differences between groups were evaluated using discriminant analysis, clustering, nonlinear estimation, structural equation modeling, Monte Carlo, bootstrap simulation and neural networks computing. Results: It was revealed that early detection of GC from HD (n = 256) significantly depended on: leucocytes (abs, total), segmented neutrophils (%, abs, total), lymphocytes (%), monocytes (%, abs, total), stick neutrophils (%, abs, total), eosinophils (%, abs, total) (P = 0.007-0.000). Neural networks computing, genetic algorithm selection and bootstrap simulation revealed relationships of early detection of GC and lymphocytes (rank = 1), segmented neutrophils (rank = 2), monocytes (3), stick neutrophils (4), leucocytes (5), eosinophils (6). Correct detection of early GCP was 100% by neural networks computing (error = 0.000; area under ROC curve = 1.0). Conclusions: Early detection of GC from HD significantly depended on blood cell circuit.

Application of Bootstrap Simulation and Asymmetric Causal Approach to Fiscal Deficit-Inflation Nexus

This paper investigates the symmetric and asymmetric relationship between fiscal deficits and inflation in Nigeria within the context of bootstrap simulations with leverage adjustments using the quarterly frequency data from 1981Q1 to 2016Q4. The findings reveal that there is neither symmetric nor asymmetric causality between fiscal deficits and inflation in Nigeria. This implies that the fiscal deficits in Nigeria are not inflationary; and also, that persistent double-digit inflation rates are not the causal agents spurring perennial increase in fiscal deficits in Nigeria. This study, therefore, concludes that fiscal deficits could be used to stimulate output level in Nigeria without fueling inflationary spiral in the economy. JEL Classification: C32, E17

Learn Smart with Less: Building Better Online Decision Trees with Fewer Training Examples

Online decision tree models are extensively used in many industrial machine learning applications for real-time classification tasks. These models are highly accurate, scalable and easy to use in practice. The Very Fast Decision Tree (VFDT) is the classic online decision tree induction model that has been widely adopted due to its theoretical guarantees as well as competitive performance. However, VFDT and its variants solely rely on conservative statistical measures like Hoeffding bound to incrementally grow the tree. This makes these models extremely circumspect and limits their ability to learn fast. In this paper, we efficiently employ statistical resampling techniques to build an online tree faster using fewer examples. We first theoretically show that a naive implementation of resampling techniques like non-parametric bootstrap does not scale due to large memory and computational overheads. We mitigate this by proposing a robust memory-efficient bootstrap simulation heuristic (Mem-ES) that successfully expedites the learning process. Experimental results on both synthetic data and large-scale real world datasets demonstrate the efficiency and effectiveness of our proposed technique.