A Generalized Multiple Criteria Data-Fitting Model With Sparsity and Entropy With Application to Growth Forecasting

In this article, we present an extended data-fitting model which involves different and conflicting criteria, propose algorithm based on a scalarization technique to solve it. Our integrates in unique framework three namely, term, the entropy sparsity of set unknown parameters. This can be analyzed by means multiple criteria decision-making techniques. We then validate proposed modified using two computational experiments: analyze problem handwritten digit recognition logistic regression deep neural network model, respectively. final part employ methodology forecasting instead. Given importance techniques predict future, turn lead positive impacts firm performance, numerical experiments focusing forecast US GDP. first one, proceed iterated function system with grayscale maps-type fractal operator, and, second implement network-based model.