Anonymized Dataset for Predicting Financial Reporting Timeliness in a Family Firm
Description
This dataset contains 18 monthly observations of anonymized, engineered variables used to model financial reporting timeliness in a family-owned firm. The study is titled : Predicting Financial Reporting Timeliness with ERP Data: A Machine Learning Case Study of the Governance Bottleneck in a Family-Owned Firm, using Python algorithms.
Files
Steps to reproduce
"The dataset contains the final, anonymized variables used for the logistic regression model in the our paper. The target variable is 'DELAY' (a binary indicator). The predictor variables are 'TOTAL_TRANSACTION_VALUE', 'MANUAL_ENTRY_RATIO', and 'MANAGEMENT_REVIEW_LAG'. To reproduce the main findings, standardize the predictor variables and fit a logistic regression model to predict 'DELAY'. The full methodology is detailed in the associated publication."
Institutions
- Universite du 20 aout 1955 de Skikda