Econometric analysis of factors affecting oil well profitability under limited data sample conditions

Well profitability is a key performance indicator of oil-producing enterprises, reflecting the influence of production, technological, and economic factors on financial results. The use of econometric methods in analyzing individual wells is complicated by limited data and high correlations among explanatory variables, leading to multicollinearity and reduced data reliability. A review of domestic and foreign studies shows that existing works focus mainly on improving regression analysis or examining individual efficiency factors, while the procedure for constructing an econometric model under small-sample conditions remains underexplored. The aim is to develop and test an algorithm for econometric analysis of oil well profitability factors, ensuring a statistically robust and economically interpretable model with limited observations. The empirical basis comprises data from 18 oil wells of one enterprise. The dependent variable was well profitability; explanatory factors included oil production volume, water cut, fixed and variable costs, and the mineral extraction tax. Correlation analysis, simple and multiple linear regression, multicollinearity diagnostics, graphical residual analysis, and R software were used. The research produced an econometric modeling algorithm encompassing sequential diagnostics of factor relationships, multicollinearity detection, model specification refinement, and re-estimation of the regression. Removing a multicollinear factor improves statistical stability and enables economically sound interpretation of individual indicators’ impact on well profitability. Practical relevance lies in the algorithm’s applicability by oil-producing companies for analyzing profitability determinants of individual wells. A limitation is the small sample size, indicating the need for further validation on larger production datasets.