General Regression Neural Network for Software Effort Estimation of Small Programs Using a Single Variable

Software development effort estimation always remains a challenging task for project managers. New techniques are applied to estimate effort. Predicting effort for small programs in educational setting is a difficult task. Minimum number of independent variables should be used to reduce data collection effort. Evaluation of accuracy is a major activity as many methods are proposed in the literature. Here, we have applied General Regression Neural Network (GRN) and compared the results with Linear Least Squares Regression (LSR) for one and two independent variables. Results are evaluated using statistical tests and effect size. The results show that accuracy of GRN and LSR with one and two variables are not different for small programs.