2026-10-12T20:23:27Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34049683 2026-10-01T17:52:04Z category_13 category_734 category_931 category_811 category_128 category_132 portal_5 item_type_3 month_year_10_2026

<p>Feature Partitioning and Definitions.</p> Xiaojing Fan (464555) Genetics Biological Sciences not elsewhere classified Information Systems not elsewhere classified Mathematical Sciences not elsewhere classified Hematology Infectious Diseases revised classifier specification reusable audit protocol raw ctgan failed patterns later reported observed marginal effects layer neural network 30 group splits worsened dependence fidelity risk method development guided simulation framework div >< p administrative audit data enforces specified income level accounting identities xgboost achieved 0 auc remained 0 risk data xgboost roc sheet identities fidelity trade administrative records guided generators test data synthetic data year panel uniformly small test whether tdgm ), tax burden synthetic tax synthetic benchmarks support tax study uses stock generation sensitive metrics selected flows sectional comparison risky observations reproducible benchmark real tax profit margin pricing indicator internal properties induced information implementation reproduces heterogeneous rather guided simulator generating process gap ratio explicit measurement <div><p>Synthetic benchmarks can support tax-risk method development when administrative records are inaccessible, but a simulator can also predetermine the patterns later reported as detection findings. We present the Tax Data Generation Model (TDGM), a mechanism-guided simulator that enforces specified income-statement and balance-sheet identities, generates a three-year panel for 10,000 firms, and injects four prespecified risk strategies. This study uses no administrative audit data; its purpose is to test whether the implementation reproduces its design assumptions and to provide a reproducible benchmark. All predictive evaluations used five repeated 70/30 group splits by firm, so no firm appeared in both training and test data. Observed marginal effects were heterogeneous rather than uniformly small: for all risky observations, Cohen’s was −1.29 for profit margin, −1.20 for tax burden, 1.05 for the transfer-pricing indicator, and 0.15 for the income-gap ratio. In a controlled cross-sectional comparison, raw CTGAN failed both row-level accounting identities, whereas deterministic projection restored 100% validity but worsened dependence fidelity. Under the revised classifier specification, a one-hidden-layer neural network with conventional L2 regularization achieved mean ROC-AUC 0.9942 and PR-AUC 0.9710; XGBoost achieved 0.9714 and 0.8975, respectively. Removing the income-gap ratio changed XGBoost ROC-AUC only from 0.9714 to 0.9702, while restricting the model to raw Level 1 variables reduced it to 0.8941. At 1% prevalence, XGBoost ROC-AUC remained 0.9720 but PR-AUC fell to 0.6703, showing why prevalence-sensitive metrics are required. Together, the three contributions are a constraint-by-construction simulator, a reusable audit protocol for mechanism-guided generators, and explicit measurement of the accounting-repair/statistical-fidelity trade-off. Because selected flows are changed after stock generation, derived ratios can carry design-induced information. These are internal properties of a synthetic data-generating process, not estimates of real tax-evasion behavior or deployable audit performance.</p></div> 2026-10-01T17:51:46Z Dataset Dataset 10.1371/journal.pone.0353137.t002 https://figshare.com/articles/dataset/_p_Feature_Partitioning_and_Definitions_p_/34049683 CC BY 4.0