Regression Diagnostics with Predicted Residuals of Linear Model with Improved Singular Value Classification Applied to Forecast the Hydrodynamic Efficiency of Wave Energy Converters

In the preliminary stages of design oscillating water column (OWC) type wave energy converters (WECs), we need a reliable cost- and time-effective method to predict hydrodynamic efficiency as function parameters. One cheapest approaches is create multiple linear regression (MLR) model using an existing data set. The problem with this approach that reliability MLR predictions depend on validity assumptions, which are either rarely tested or sub-optimal procedures. We offer series novel methods for assumption diagnostics apply in our case study prediction hydrodynamics OWC WECs. Namely, propose: procedure identification zero singular values matrix; modified algorithm stepwise regression; detect heteroskedasticity identify statistically significant but practically insignificant heteroscedasticity original model; test nullity assumption; Jarque–Bera Monte Carlo error normality test. study, deviations from assumptions classical normal were fully diagnosed dealt with. newly proposed algorithms based improved value decomposition (SVD) matrix predicted residuals successfully new family goodness-of-fit measures. empirically investigated correct placement elaborate outlier detection overall diagnostic sequence. As result, constructed design. useful tool at can produce highly WEC performance provided constructing procedures reflect latest advances statistics. main advantage models compared other modern black box their known be practice, increases predictions.